WVS For Multi Platforms v5.1 Product Documentation

WebSphere Voice Server
Version 5.1.3
WebSphere Voice Server for Multiplatforms
Tenth Edition (April 2006) This edition applies to IBM WebSphere Voice Server version 5, release 1.3. This edition applies to all subsequent releases and modifications until otherwise indicated in new editions. A form for readers comments appears at the back of this publication. If the form has been removed, address your comments to: International Business Machines Corporation Department MMOA P.O. Box 12195 Research Triangle Park, North Carolina 27709-2195 When you send information to IBM, you grant IBM a nonexclusive right to use or distribute the information in any way it believes appropriate without incurring any obligation to you. Copyright International Business Machines Corporation 2005, 2006. All rights reserved. US Government Users Restricted Rights Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.
Contents
Chapter 1. Product overview . . . . . . 1
Speech in e-business . . . . . . . . . . . Brief introduction to speech . . . . . . . . Product description . . . . . . . . . . . Features . . . . . . . . . . . . . . . Accessibility . . . . . . . . . . . . Administrative ease of use . . . . . . . . Compatibility with IVRs and gateways . . . Continuous access . . . . . . . . . . Scalability . . . . . . . . . . . . . Rapid problem determination . . . . . . . Security . . . . . . . . . . . . . . Product GUIs and tools . . . . . . . . . . Relationship to other versions of WebSphere Voice Server products . . . . . . . . . . . . WebSphere platform and related software . . . About this information . . . . . . . . . Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2 4 6 6 7 8 8 9 9 9 9 Installing interim fixes and fix packs for the product . . . . . . . . . . . . . . . Installing deployment manager . . . . . . . Installing WebSphere Voice Server console extensions on a deployment manager . . . . . Installing the WebSphere Edge Component: Load Balancer . . . . . . . . . . . . . . Customizing the options response file . . . . Installation: Resources for learning . . . . . Installation tips . . . . . . . . . . . . Using the installation verification test. . . . . . Installation verification test command . . . . firststeps command . . . . . . . . . . . Installing the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server . . . . . . Prerequisites for the IBM TTS Connector for Genesys Voice Platform . . . . . . . . . Uninstalling the product . . . . . . . . . . Uninstalling TTS and ASR languages . . . . . Uninstalling Network Deployment . . . . . Uninstalling base WebSphere Application Server Manually uninstalling WebSphere Application Server . . . . . . . . . . . . . . . Uninstalling the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server . . . . 79 80 81 82 83 83 83 83 85 85 86 88 88 89 90 90 90 91
11 12 13 15
Chapter 2. Developing . . . . . . . . 17
Standards for developing . . . . . . . . . Understanding VoiceXML . . . . . . . . Developing a speech application . . . . . . Testing grammars . . . . . . . . . . . Tuning speech resources for effective applications Preparing a lexicon of ASR pronunciations . . Canadian French phonetic pronunciation . . German phonetic pronunciation . . . . . Japanese phonetic pronunciation . . . . . Latin American Spanish phonetic pronunciation Simplified Chinese phonetic pronunciation . . UK English phonetic pronunciation . . . . US English phonetic pronunciation . . . . . . . . . . . . . . . 17 17 18 18 19 20 20 22 24 26 27 38 39
Chapter 5. Configuring . . . . . . . . 93
Configuring using the administrative console . . . 93 Configuring ASR settings . . . . . . . . . 93 Configuring TTS settings . . . . . . . . . 94 Configuring using scripting and command line tools 95 Adding a voice server . . . . . . . . . . 96 Removing a voice server . . . . . . . . . 99 Creating custom scripts for configurations . . . 101 Configuration Service MBean . . . . . . . 102 Configuring default and available TTS voices 104 Configuring a MIME map on your customer Web server . . . . . . . . . . . . . . . . 105 Configuration parameters . . . . . . . . . 105 Configuring WebSphere Voice Server to support Genesys . . . . . . . . . . . . . . . 112 Configuring WebSphere Voice Server to support WebSphere Voice Response . . . . . . . . . 112 Configuring the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server . . . 113
Chapter 3. Planning . . . . . . . . . 43
Understanding the speech application environment Estimating resources for ASR engines . . . . . Estimating resources for TTS engines . . . . . Estimating resources for a LAN . . . . . . Identifying hardware and software requirements . Compatibility with IVRs and gateways . . . . Single machine topology . . . . . . . . . Multiple machine topology . . . . . . . . Establishing multiple machine environments . . . . . . . . . . 43 47 47 48 48 50 50 53 54
Chapter 4. Installing . . . . . . . . . 57
Installing WebSphere Voice Server . . . . . WebSphere Voice Server package . . . . Installing the product . . . . . . . . . Prerequisites . . . . . . . . . . . Using the LaunchPad to start the installation Installing version 5.1.2 and earlier versions . Installing version 5.1.3 . . . . . . . . Installing silently . . . . . . . . . Installing additional languages . . . . .
Copyright IBM Corp. 2005, 2006
. . . . . . . . .
. . . . . . . . .
57 58 59 60 61 62 68 76 78
Chapter 6. Administering . . . . . . 117

Using the administrative console . . . . . . . Starting and stopping the administrative console Login settings . . . . . . . . . . . . Administrative console areas and icons . . . . Filtering on the administrative console . . . . Monitoring voice servers and resource status Starting voice servers . . . . . . . . . . Stopping voice servers . . . . . . . . . 117 117 119 120 121 122 125 125
iii
Security . . . . . . . . . . . Managing using scripting . . . . . . getParam sample . . . . . . . . putParam sample . . . . . . . . startVoiceServer sample . . . . . . stopVoiceServer sample . . . . . . statusVoiceServer sample . . . . . backupVoiceServerConfig sample . . . restoreVoiceServerConfig sample . . . Automatically restarting server processes . Starting and stopping the IBM HTTP Server
. . . . . . . . . . .
. . . . . . . . . . .
. . . . . . . . . . .
125 126 127 128 129 130 131 132 134 135 142
Chapter 7. Tuning
. . . . . . . . . 145
. . . . . . 145 145 146 147 147 148
Tuning your configuration . . . . . . . . Tuning WebSphere Voice Server . . . . . Tuning WebSphere Application Server . . . Tuning AIX . . . . . . . . . . . . Tuning Linux: Red Hat 3.0 and SUSE 8.0 . . Tuning threading on SUSE 8.0 with SP2a . . Tuning Windows Server 2003 Enterprise and Standard editions . . . . . . . . . . Tuning the Cisco gateway . . . . . . .
. 149 . 150
Using IBM Support Assistant . . . . . . . . Troubleshooting by component: What is not working? . . . . . . . . . . . . . . . Installation troubleshooting tips . . . . . . Administration and administrative console troubleshooting tips . . . . . . . . . . Security troubleshooting tips . . . . . . . IBM HTTP Server troubleshooting tips . . . . TTS troubleshooting tips. . . . . . . . . ASR troubleshooting tips . . . . . . . . IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server troubleshooting tips Gateway communication problems . . . . . voiceTest troubleshooting tips . . . . . . . Troubleshooting by task: What are you trying to do? . . . . . . . . . . . . . . . . . Troubleshooting installation problems . . . . Troubleshooting testing and first time run problems . . . . . . . . . . . . . . Troubleshooting application runtime and management problems . . . . . . . . .
460 460 461 461 462 462 463 467 473 474 474 475 475 476 477
Chapter 9. Resources for learning . . 479 Chapter 10. Notices and trademarks 481
Chapter 8. Troubleshooting . . . . . 153

Messages and alarms . . . . . . . Monitoring log messages . . . . . Messages by component . . . . . Identifier . . . . . . . . . . Enabling and monitoring trace . . . Collecting information for IBM Support Running the collector tool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 154 154 452 453 458 458
Glossary . . . . . . . . . . . . . 483 Index . . . . . . . . . . . . . . . 491
iv
Chapter 1. Product overview

These topics describe the product and its surrounding business and technology context.
Speech in e-business
Speech technology has come of age, and IBM WebSphere Voice Server draws on more than 30 years of worldwide speech research and technology expertise to bring you state-of-the-art functionality. What does speech offer your e-business? Simply put, speech technology offers customer satisfaction. You have heard the common approach to telephone-based customer service: For x, press 1. For y, press 2. For z, press 3. This clunky form of interaction relies on keypad presses and limited numerical user input, creating a confusing maze of menus and a less-than-satisfactory customer experience. At the same time, using live operators for customer service results in inconsistent service quality and significant expense. Speech technology is the future of e-business because it enables more natural, intuitive, and engaging customer service for less cost. Implementing speech technology means your business is always accessible to your customers. Speech applications often provide return on investment in as few as three to six months. Speech technology means that your customer service representatives can handle the complex service tasks for which they are best equipped, instead of simple, repetitive customer transactions. With speech, simple customer service needs might be handled by technology to promote a more cost-effective customer service strategy in your e-business. In addition, a well-designed speech user interface can improve customer satisfaction and loyalty over Web site self-service by giving your customers the flexibility to receive their service anywhere, anytime from their telephone, the most common personal technology. The numerous benefits of speech technology for e-business include: v Interaction with callers is easier and more natural. v Menus can be eliminated or flattened, for more subtle and intuitive navigation. v Call durations can be minimized, meaning less cost per transaction. v Interaction with customers can occur 24 hours a day, 7 days a week. v Customers interact with your business using their telephone or cellular telephone, resulting in continuous access to the customer base regardless of their location. v Individuals with physical or perceptual disabilities might have greater access to e-business services. v Enterprise branding is extended to a new channel phone-based interaction. v Overall customer service costs are decreased. v Return on investment for speech application development often occurs in as few as three to six months.
v Opportunities for integrating and streamlining business processes arise as speech applications are developed. v New business opportunities can arise using speech technology. These benefits combine to increase customer satisfaction and loyalty, meaning that speech technology is central to effective, efficient e-business on demand.
Brief introduction to speech

This section is a quick primer on speech technology. Use it as an initial reference to help you get started with speech technologies and integrate this advanced form of interaction in your business environment. Speech refers to the recognition and production of spoken language. Speech technology, at its most basic, includes one technology for recognition, known as automatic speech recognition (ASR), and a second technology for the production of spoken language. Technology-generated speech, or the system voice, is achieved in one of three ways: v Using text-to-speech (TTS) technology, which creates spoken language from written language. v Using recorded human voice segments (prompts). v By combining TTS and prerecorded voice segments. Currently, combinations of TTS and recorded voice segments provide the most flexible and cost-effective solutions for the majority of speech applications. The accuracy of ASR (did it accurately recognize what the user said?) and the pleasantness and naturalness of the system voice largely determine the quality of speech applications. Other elements of quality are determined by the speech interface design of the application, such as the flow of the dialogue and use of language in prompts.
How does ASR work?

ASR enables a server to recognize general, naturally flowing utterances from a wide variety of users. It recognizes the callers answers and uses that input to facilitate the flow of the call. ASR technology converts acoustic waves from a users spoken words to a form that is usable by a speech application. This means that speech input can be accepted from a user and processed in the same way as dual tone multi frequency (DTMF) input, text, or numeric data. The outcome of a correctly recognized user utterance is determined by the call flow of a speech application. Several important terms are associated with ASR: Voice model A set of parameters for the language to be recognized, including the acoustic model, the numeric model of how the language sounds when spoken. Grammar The specification of the words and phrases that are active and can be recognized by the ASR engine at a given time. Grammars can be created dynamically or at application development time, and can be loaded, unloaded, activated and deactivated under the voice application control.
Lexicon The set of words in a given language or system. A lexicon refers to the set of files that comprise the vocabulary of an application or system. Lexicons are used by the ASR engine to increase its recognition accuracy. See Preparing a lexicon of ASR pronunciations for a sample lexicon.
How does TTS work?

TTS is typically used as part of the system voice in speech applications. It often is used for variable items in a dialogue, such as a user address or name, specific dialogue items that are dynamic and thus cannot be predicted by the application developer. In a working speech interaction, TTS segments can be spliced into prompts spoken by a professional voice talent to create unique messages for each user. For example, the sentence Your balance is 7513 dollars and 45 cents can use a recorded voice talent for the static your balance is and TTS for 7513 dollars and 45 cents because the amount of money in each users account is dynamic information and too variable to record professionally in a cost-effective way. Speech synthesis can be controlled through the use of the following: Speech Synthesis Markup Language (SSML) A markup language application developers use to create TTS standardized synthesizable output from speech engines. SSML allows users to control the synthesis by modifying pronunciations, prosody, text processing, and normalization. For more information on SSML, refer to the W3C recommendation for the specification. Lexicon Lexicons contain pronunciation information which can be used to override the TTS engines automatically derived pronunciations. Lexicons specified in an SSML document are active during the synthesis of that SSML only. Lexicons containing a large set of word pronunciations needing to remain active should be preloaded. For more information regarding preloaded lexicons, see the Configuring TTS section of the Information Center. For information about lexicon syntax, refer to Pronunciation Lexicon Markup for the W3C Speech Interface Framework.
What does WebSphere Voice Server do?

WebSphere Voice Server provides the ASR and TTS resources that enable speech-based interaction. Several system-user-system responses make up a conversational interaction. A speech application or VoiceXML content, provides the logic and flow of dialogue. A VoiceXML browser provides access to the speech application or VoiceXML content from a Web server. The VoiceXML browser interprets the application, processing conversational interactions requiring ASR and/or TTS resources. When these elements are combined, the speech interface is perceived as a series of system utterances that require user responses. The WebSphere Voice Server allows system administrators to configure, monitor, and troubleshoot voice server resource usage in deployed speech applications. Related information WebSphere Voice family Speech Synthesis Markup Language Version 1.0
Speech Recognition Grammar Specification Version 1.0
Product description
IBM WebSphere Voice Server offers a Media Resource Control Protocol (MRCP) server that conforms to draft four of the MRCP version 1 specification definition for speech technology.

IBM WebSphere Voice Server confirms IBMs commitment to support, adopt, and drive open standards. With WebSphere Voice Server, IBM moves from providing proprietary application program interfaces (APIs) to its speech technology to the industry standard the MRCP interface. WebSphere Voice Server also builds on the base of other IBM WebSphere products to provide scaling, load balancing, failover, recovery, systems management, logging, tracing, and problem determination consistent with the IBM WebSphere family. WebSphere Voice Server leverages the ease of use and administration capabilities of IBMs WebSphere Application Server, the leading application server in the industry, with proven capabilities for scaling and performance. For additional information on WebSphere Application Server, see the WebSphere Application Server 5.1.x Information Center.
ASR support
Automatic speech recognition (ASR) is the technology that matches spoken input to the expected input (grammar) in a speech application. WebSphere Voice Server offers eight languages with significantly improved performance in noisy environments, especially in reducing false barge-ins. These languages include: v Australian English v Canadian French v German v Japanese v Latin American Spanish v Simplified Chinese v U.K. English v U.S. English
TTS support
TTS converts written text into spoken words and often provides the voice of a speech application. WebSphere Voice Server provides improved synthesis and signal processing algorithms. TTS supports: v Speech Synthesis Markup Language (SSML) v SSML lexicons v Persistent lexicons TTS also offers several advanced TTS male and female voices for the following languages: v Australian English (The TTS for Australian English uses the UK English TTS) v Canadian French (one male voice, one female voice)
German (one male voice, one female voice) Japanese (one male voice, one female voice) Latin American Spanish (one male voice, one female voice) Simplified Chinese (one female voice) U.S. English (one male voice, two female voices, two fifth generation female voices) v U.K. English (one male voice, two female voices, one fifth generation female voice) v v v v v Restriction: v TTS supports adult voices only. v Fifth generation voices are smoother and more natural sounding, but also larger. Depending on the capacity of your WebSphere Voice Server, the number of voices that can be loaded might be limited to one fifth generation voice and an additional voice. To install and use the fifth generation voices, install fix pack 1 for WebSphere Voice Server 5.1.3 and then install the fifth generation voices you plan to use. Download the fix pack and voices from the WebSphere Voice Server support site.
Multiple language support

WebSphere Voice Server now supports multiple languages on one server and multiple language applications within a session. All of the supported ASR and TTS languages are available for this new feature.
MRCP server
The Internet Engineering Task Force (IETF) defines MRCP as a protocol for controlling media service resources, such as a speech recognizer or a speech synthesizer. The following figure shows the relationship of WebSphere Voice Server to a partner platform and the application server.
Figure 1. The MRCP environment
The WebSphere Voice Server base provides the features of the MRCP server that conform to draft four of the MRCP version 1 specification definition. When you add a language support package, the MRCP server acts as a recognizer, providing a standard interface into the ASR resources. The Concatenative TTS package supports the synthesizer functions defined in the MRCP specifications. When you deploy WebSphere Voice Server, you can install both recognition and synthesis resources on the same physical server, or on separate servers. Scaling of the MRCP server is available using the WebSphere Load Balancer. The Load Balancer enables multiple physical servers to present a single server name to which the MRCP clients can connect.
Features
By leveraging the features and functions that an administrator uses for an application server, WebSphere Voice Server provides administrative ease of use to voice server administrators. For e-businesses, speech allows an alternate means of access to business services and transactions, especially for individuals with disabilities. Using an architectural foundation provided by WebSphere Application Server, IBMs premier product for e-business technology, the WebSphere Voice Server also provides extensive administrative and performance benefits for complex speech technology architectures, including the following.
Accessibility
Accessibility features enable users with physical disabilities, such as restricted mobility or limited vision, to use software products.
IBMWebSphere Voice Server for Multiplatforms is a voice application enabler, which appeals to the major user groups impacted by speech technology system administrators, application developers, and application users. WebSphere Voice Server provides the following accessibility features for each user group: System Administrators v Provides accessible administration and monitoring interfaces for speech technology resources. v Offers text through standard system function calls or through an open application programming interface (API), which supports interaction with assistive technology. v Provides accessible documentation. Telephone Callers (users) v Offers the capability of providing people with visual, manual, and or mobility impairments access to business data and services using auditory user interfaces. v Helps businesses comply with Section 508 of the Americans with Disabilities Act. Application Developers v Provides an accessible development environment for speech applications. The administrative console is the primary interface for interacting with the product. This console displays within a standard Web browser. By using an accessible Web browser, such as Microsoft Internet Explorer, administrators are able to: v Use screen-reader software and a digital speech synthesizer to hear what is displayed on the screen. v Use voice recognition software to enter data and to navigate the user interface. v Operate features by using the keyboard instead of the mouse. You can also configure and administer product features by using standard text editors and scripted or command line interfaces instead of the graphical interfaces provided. When appropriate, the documentation for specific product features contains additional information about the accessibility of the features.
Administrative ease of use

One feature of WebSphere Voice Server is its administrative console, a GUI control panel for administration and monitoring of speech resources. The console provides views of overall system status and detailed status of automatic speech recognition (ASR) and text-to-speech (TTS) resources, as well as general resources (system status, free memory). System administrators can obtain at-a-glance knowledge of their system by reviewing a single interface. In addition to a speech-centric view of voice servers, administrators familiar with WebSphere Application Server can also view their system from the Application Server perspective, for comprehensive system understanding.
The console is also a centralized configuration tool. Administrators can use the console panel to configure the ASR and TTS engines, preloaded grammars, lexicons, and installed voices. Finally, for administrators who prefer to administer using scripts, WebSphere Voice Server provides several example Jacl scripts to help you quickly learn the skills to produce your own unique administration code.
Compatibility with IVRs and gateways

For a listing of the current compatible Interactive Voice Response (IVR) systems and gateways for WebSphere Voice Server, see IVR and gateway compatibility for WebSphere Voice Server for Multiplatforms on the Web.
Continuous access
Your speech applications provide customer service and business transactions, save money, and streamline e-business processes. WebSphere Voice Servers ability to support complex architectures ensures that your business will always be available to your customers. Availability and failover support Avoid a single point of failure and maximize system availability by ensuring that the topology has some degree of process redundancy. High-availability topologies typically involve horizontal scaling across multiple machines. Load balancing capability A Load Balancer server performs intelligent load balancing to determine where to send a TCP/IP request. It can direct client RTSP/MRCP requests to available voice servers, bypassing any that are offline. Another server can back up the Load Balancer server, to eliminate it as a single point of failure. Workload management of voice servers also improves availability and failover support. Failover support distributes client RTSP/MRCP requests to the remaining servers, which ensures continued client access without significant interruptions. Maintainability The ability to update hardware and software is another important consideration for many speech technology infrastructures. WebSphere Voice Server allows you to create multiple Load Balancer cells (horizontal scaling) that can make a system easier to maintain because you can take individual machines offline without interrupting other machines that are running a speech application. Although taking a system offline in a cell can reduce overall throughput, for those applications that must continue running in the event of maintenance, WebSphere Voice Servers flexibility allows you to design your system appropriately. Dynamic cell management Modifying a Voice Server cell deployment without interrupting its operation enhances the manageability and flexibility of the overall system. For instance, administrators can add or remove cell members or nodes, to handle variations in the client load, change server characteristics, propagate changes to its cell members, and temporarily stop servers for maintenance.
Scalability
Adding more machines increases processing power, which is important for rapid exchanges in speech applications. WebSphere Voice Server provides horizontal scalability, adapting readily to a greater or lesser intensity of use, volume, or demand. Horizontal scaling creates additional WebSphere Voice Server processes on multiple physical machines to take advantage of additional processing power. You can use WebSphere Application Server Edge (Load Balancer) with horizontal scaling to provide a single system view of multiple physical machines.
Rapid problem determination

WebSphere Voice Server includes comprehensive messages and alarms to keep you fully informed about system status so that you can quickly target and resolve problems using the administrative console. All trace and log messages are now in a centralized repository. Troubleshooting tools, including Tivoli Performance Viewer, enable real-time analysis and system monitoring. If you require support, a Collector Tool gathers system information for support personnel. The new Voice Trace Analyzer, available in the latest release of WebSphere Voice Toolkit, lets you examine recognition data from a WebSphere Voice Server system only. It can read multiple trace.log files to build a comprehensive overview of your systems recognition performance and improve the awareness of recognition problems. All of these features expedite problem determination and resolution, saving you time and money.
Security
IBM WebSphere Voice Server uses the security infrastructure and mechanisms provided by WebSphere Application Server. WebSphere Application Server allows administrators to protect a server using two security features: Global Security and Java 2 Security. These security features are managed through the Security panel on the administrative console. Important: Java 2 Security cannot be enabled in order for WebSphere Voice Server to function properly. If you need security, enable Global Security in the WebSphere Voice Server administrative console. Do not select the option to enforce Java 2 Security. Before you enable global security read the technote about using Local operating system user registries. For detailed information about the WebSphere approach to security, refer to Welcome to Security in the WebSphere Application Server Information center.
Product GUIs and tools

IBM WebSphere Voice Server provides a variety of graphical and non-graphical user interfaces, to provide an end-to-end speech solution for your e-business.
Tools for installing

WebSphere Voice Server includes easy-to-use tools to get your voice server installed and running with maximum ease of use. Launch Pad A graphical interface for starting the product installation. It also provides links to helpful install information. See Using the LaunchPad to start the installation on page 61. Installation wizard A graphical interface that leads you through the WebSphere Voice Server installation in a series of steps. First Steps A comprehensive beginning point from which you can access the information center, test your installation, and start or stop the WebSphere Voice Server. See the topic on the firststeps command on page 85 for more information.
Tools for developing applications

WebSphere Voice Toolkit WebSphere Voice Toolkit offers a full-function development environment for VoiceXML applications, making it an important part of any successful speech development effort. The Voice Toolkit includes the Communication Flow Builder, which allows rapid development of your speech applications for usability evaluation and iterative design. The latest release of Voice Toolkit also includes the Voice Trace Analyzer. For more information on the different releases of Voice Toolkit, see the WebSphere Voice Toolkit product page. voiceTest The voiceTest tool provides you assistance as you develop speech applications and prepare them to work with WebSphere Voice Server. The voiceTest tool is a command-line interface that allows basic interaction with the WebSphere Voice Server MRCP server. The tool is used for installation verification, grammar testing, and ASR and TTS engine testing.
Tools for troubleshooting

WebSphere Voice Server also provides tools to help you with rapid problem determination for your voice system. Administrative console The administrative console is a graphical interface for performing deployment and system administration tasks. For more information, see Using the administrative console on page 117. Collector Tool The Collector Tool gathers information about your WebSphere Voice Server installation (including WebSphere Application Server) and packages it in a JAR file that can be sent to IBM Customer Support to assist in problem determination and analysis. For more information, see Collecting information for IBM Support on page 458. Tivoli Performance Viewer Tivoli Performance Viewer is a GUI that allows you to monitor system performance in real time. This useful tool can be used with the
10
moment-in-time monitoring provided by the WebSphere Voice Server administrative console. For more information, see Tivoli Performance Viewer. Voice Trace Analyzer The latest version of WebSphere Voice Toolkit includes the Voice Trace Analyzer, which lets you examine recognition data from a WebSphere Voice Server system only. It can read multiple trace.log files to build a comprehensive overview of your systems recognition performance and improve the awareness of recognition problems.
Relationship to other versions of WebSphere Voice Server products

Even though the WebSphere Voice Server package includes WebSphere Voice Server version 4.2 for AIX, there is no relationship between the two product versions. WebSphere Voice Server version 5.1.x is not compatible with WebSphere Voice Server 4.2; however, you can migrate your WebSphere Voice Server version 4.2 speech applications to WebSphere Voice Server version 5.1.x. For more information on how to migrate your applications, refer to Migrating Applications from IBM WebSphere Voice Server 4.2 to 5.1.x for Use with IBM WebSphere Voice Response. Both WebSphere Voice Server versions 4.2 and 5.1.x are compatible with IBM WebSphere Voice Response. To use WebSphere Voice Response with WebSphere Voice Server version 5.1.x, you must use the WebSphere Voice Server Connector, which is included in the interim fixes for WebSphere Voice Server versions 5.1.1 and 5.1.2. The WebSphere Voice Server Connector is included with the product CDs for WebSphere Voice Server version 5.1.3. WebSphere Voice Server version 5.1.x does not include the WebSphere Voice Server Speech Technologies, WebSphere Voice Server/Windows or Cisco/Dialogic products. WebSphere Voice Server version 5.1.x supports the following operating systems: v v AIX 5L version 5.3 with maintenance level (ML) 3 Red Hat Enterprise Linux WS/ES/AS for Intel 3.0 with Update 1, Update 3, or Update 4 (2.4 Kernel) - supported in WebSphere Voice Server version 5.1.3
Red Hat Enterprise Linux WS/ES/AS for Intel 3.0 with Update 1 only (2.4 Kernel) - supported in WebSphere Voice Server versions 5.1.1 and 5.1.2 v SUSE LINUX Enterprise Server 8, powered by United Linux 1.0 (SLES-8) (Intel) with Service Pack 2a or 3 v Windows Server 2003 Enterprise or Standard editions with or without Service Pack 1 v Note: If you are using Service Pack 1, download and apply the WebSphere Voice Server interim fix, IC47845, from the product Support site. Also refer to the WebSphere Voice Server product page to find the current list of platforms supported.
11
WebSphere platform and related software

IBM provides a family of WebSphere products and other related software that create a reliable, standards based infrastructure for your e-business. WebSphere Voice Server is closely related to WebSphere Application Server, which is the architectural foundation of IBMs speech technology. To learn more about WebSphere Application Server and increase your knowledge of WebSphere Voice Servers advanced capabilities, refer to the WebSphere Application Server 5.1.x Information Center. IBMs other WebSphere products complement IBM speech technology for comprehensive e-business solutions.
Foundation and tools

In addition to WebSphere Application Server, another foundation product provides scalability and productivity for rapidly and reliably growing e-business applications: v WebSphere MQ This software exchanges information among more than 35 platforms with assured delivery. For more information, see http://www.ibm.com/software/ts/ mqseries/. v Rational Application Developer for WebSphere Software and Rational Web Developer for WebSphere Software Optimized for IBM WebSphere software and capable of supporting multi-vendor runtime environments, IBM Rational Application Developer for WebSphere Software and Rational Web Developer for WebSphere Software are powered by the Eclipse open source platform so you can adapt and extend your development environment to match your needs and increase productivity. For more information, see http://www.ibm.com/software/awdtools/developer/ application/index.html and http://www.ibm.com/software/awdtools/ developer/web/index.html.
Reach and user experience products

The following products collaborate to help you achieve customer loyalty by extending and personalizing user experiences: v WebSphere Portal Server This software accesses widespread and diverse data sources from anywhere, anytime, by anyone you allow. For more information, see http://www.ibm.com/ software/webservers/portal/. v WebSphere Everyplace This is a software infrastructure that supports mobile solutions, addressing the challenge of extending e-business applications to mobile devices. For more information, see http://www.ibm.com/software/pervasive/. v WebSphere Commerce These powerful solutions are designed to handle the broad range of challenges encountered when selling in Business-to-Business (B2B) and Business-to-Consumer (B2C) environments. For more information, see http://www.ibm.com/software/webservers/commerce/.
Business integration products
12
The following products collaborate to help you achieve business agility by integrating applications and automating business processes: v WebSphere Business Integration This software creates the nimble infrastructure needed to support the business imperatives of your dynamic enterprise. For more information, see http://www.ibm.com/software/integration/. v WebSphere MQ Integrator This software helps you to flexibly connect and integrate their assets within the enterprise and with trading partners. For more information, see http://www.ibm.com/software/integration/.
About this information

This information center is for planners, installers, administrators, developers, and others who plan to install, configure, administer, and develop for WebSphere Voice Server. The information in this information center is designed to help you design and implement your voice solution. Because every voice server deployment is unique, it is important to also refer to the WebSphere Voice Zone for links to other information about WebSphere Voice Server and related products. There are links to IBM Redbooks, articles, and much more. To ensure that you are using the latest version of the information center for WebSphere Voice Server, refer to the WebSphere Voice Zone product documentation page at http://www.ibm.com/developerworks/websphere/zones/ voice/proddoc.html. For more information about the information center, use the following links: v Conventions v Using the information center v Information center change history
Information center change history

Edition Edition 10 Date April 21, 2006 Updates Updates to the following topics: v Product description v Canadian French phonetic pronunciations v Estimating resources for TTS engines v Identifying hardware and software requirements v Configuration parameters Updates for new fifth generation female voices and updates for the Genesys Voice Platform (GVP).
13
Edition Edition 9
Date December 12, 2005
Updates Updates to the following topics: v Installing the product v Installation prerequisites Also updates for Genesys Voice Platform (GVP) and for the support of Windows Server 2003 Service Pack 1.
Edition 8
November 3, 2005
Information center refreshed with updates for the support of AIX and the installation of multiple languages on one machine. Information center refreshed with updates for the support of the latest releases of WebSphere Voice Response and WebSphere Voice Toolkit. Information center refreshed with Eclipse framework. Updates to the following topics: v Planning v Installing the product v Configuring v Tuning your configuration v Troubleshooting
Edition 7
June 14, 2005
Edition 6
May 5, 2005
Edition 5
March 11, 2005
Updates to the following topics: v Planning v Installing the product v Installation prerequisites v Developing v Configuring v Tuning your configuration v Troubleshooting
Edition 4
January 23, 2005
Updates to the following topics: v Installing the product v Installation prerequisites v Tuning your configuration v Message IDs
14
Edition Edition 3
Date November 23, 2004
Updates New topics including: v Phonetic pronunciation tables for German and Canadian French languages. Updates to the following topics: v Tuning your configuration v TTS troubleshooting tips
Edition 2
November 5, 2004
New topics including: v Compatibility with IVRs and gateways v Installing interim fixes, cumulative fixes, and fix packs v Installing the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server v Uninstalling the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server v Configuring WebSphere Voice Server to support Genesys v Configuring the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server v IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server troubleshooting tips Updates to the following topics: v Installing the product v Installation prerequisites v Tuning your configuration v Message IDs
Edition 1
July 16, 2004
Initial information center that was included on the WebSphere Voice Server Version 5.1 product CDs.
Conventions
The guide uses several typeface conventions for special terms and actions. These conventions have the following meanings.
15
Bold
Lowercase and mixed-case commands, command options, and flags that appear within text appear like this, in bold. Graphical user interface elements, except for titles of windows and dialogs and names of keys also appear like this, in bold. Variables, values you must provide, new terms, and words and phrases that are emphasized appear like this, in italic type. Commands, command options, and flags that appear on a separate line, code examples, output, and message text appear like this, in monospace type. Names of files and directories, text strings you must type, when they appear within text, names of Java methods and classes, and HTML and XML tags also appear like this, in monospace type. Information marked with this icon applies to WebSphere Voice Server version 5.1 Information marked with this icon applies to WebSphere Voice Server version 5.1 and subsequent versions. Information marked with this icon applies to WebSphere Voice Server version 5.1.1. Information marked with this icon applies to WebSphere Voice Server version 5.1.1 and subsequent versions. Information marked with this icon applies to WebSphere Voice Server version 5.1.2. Information marked with this icon applies to WebSphere Voice Server version 5.1.2 and subsequent versions. Information marked with this icon applies to WebSphere Voice Server version 5.1.3.
Italic Monospace
The following variables are used in this documentation: $WVS_ROOT is a variable for Linux and AIX that is defined during the installation of WebSphere Voice Server. Typing echo $WVS_ROOT at a command prompt shows you the path where WebSphere Voice Server is installed, such as . /opt/WebSphere/VoiceServer on Linux or . /usr/WebSphere/VoiceServer on AIX. %WVS_ROOT% is a Windows variable defined during the installation of WebSphere Voice Server. Typing echo %WVS_ROOT% at a command prompt shows you the path where WebSphere Voice Server is installed, such as C:\Program files\WebSphere\VoiceServer. WAS_ROOT is a variable representing the installation path of WebSphere Application Server. For reference, the WebSphere Voice Server default installation paths are the following: /usr/WebSphere/VoiceServer /opt/WebSphere/VoiceServer C:\Program Files\WebSphere\VoiceServer
16
Chapter 2. Developing
This document concentrates on tasks associated with WebSphere Voice Server product usage and administration. There are many sources of excellent information on developing speech user interfaces and applications. The speech development environment provided with WebSphere Voice Server is the WebSphere Voice Toolkit, which can help streamline application design activities. Although application development activities are typically not performed by the same professionals as voice server administration, use the sections below for an overview of the tasks involved in developing.
Standards for developing

This topic describes the supported standards for grammars and text-to-speech (TTS). The following specifications are supported for developing grammars: v Speech Recognition Grammar Specification (SRGS) 1.0 http://www.w3.org/ TR/speech-grammar/ Note: WebSphere Voice Server SRGS supports the following meta values for the http-equiv attribute: Last-Modified, Expires, and Date. Valid values for these attributes are specified in RFC 2616. v Semantic Interpretation for Speech Recognition (SISR) W3C Working Draft 1 April 2003 http://www.w3.org/TR/2003/WD-semantic-interpretation20030401/ WebSphere Voice Server TTS supports the following markup language: v Speech Synthesis Markup Language (SSML) 1.0 http://www.w3.org/TR/ speech-synthesis/
Understanding VoiceXML
WebSphere Voice Server is a Media Resource Control Protocol (MRCP) version 1 draft 4 server. Thus, Voice Extensible Markup Language (VoiceXML) browsers that support the MRCP version 1 draft 4 standard can be used with WebSphere Voice Server. A VoiceXML browser is part of a comprehensive speech solution provided with the gateway. VoiceXML is the industry-standard markup language for interactive speech applications. With VoiceXML, you can use a Web-centric programming model and a markup language specially designed to bring the speed and flexibility of Web-based development and content delivery to interactive speech applications. In other words, development skills in HTML transfer to VoiceXML, improving the time to code speech applications. VoiceXML dialogs provide a flexible presentation layer on top of the logic and data layers of your business applications. Using VoiceXML, you can build sophisticated dialogs that enable the caller to use either speech or key-presses to provide input, and to hear responses as either synthesized or prerecorded speech. The speech
17
input from callers can be either recognized or recorded. VoiceXML enables you to build usable and flexible voice interfaces to your business. In addition to accessing your business data, VoiceXML applications can also access telephony information such as called and calling number and perform simple call transfers. The recommended approach is to develop dialogs using VoiceXML in the complete solution. This can include host data accessed through Enterprise JavaBeans (EJBs) and JavaServer programs (JSPs).
Developing a speech application

The development environment provided with the WebSphere Voice Server is the WebSphere Voice Toolkit. The features of this programming environment include the following: Development environment (IDE) Communication Flow Builder tool VoiceXML editor Call Control Extensible Markup Language (CCXML) editor Editors for creating grammars written in Speech Recognition Grammar Specification (SRGS) XML or ABNF v Editors for testing grammars using speech, text, or enumeration v v v v v The toolkit also includes a pronunciation builder, an audio recorder, and a set of VoiceXML reusable dialog components. For more information on the tools provided in WebSphere Voice Toolkit, see the WebSphere Voice Toolkit Information Center. A comprehensive overview of speech application and interface design is provided in the VoiceXML Programmers Guide, which is also a related document in the WebSphere Voice Toolkit Information Center. Tip: Chapter 2 of the VoiceXML Programmers Guide, Designing a Speech User Interface, highlights the importance of tuning your application grammars and timeouts before deploying into production. Proper tuning of grammars and timeouts can minimize application errors and speed the production cycle. Also reference Using IBM Text-to-Speech Technology and Speech Synthesis Markup Language for information about using Speech Synthesis Markup Language (SSML) to incorporate IBM Text-to-Speech (TTS) technology into applications.
Testing grammars
When developing a speech application, you must also test your grammars. WebSphere Voice Server provides two different tools to test grammars the WebSphere Voice Toolkit and the voiceTest tool. The Voice Toolkit provides three GUI utilities for testing: v A grammar test tool that interfaces with a remote WebSphere Voice Server v Pronunciation migration tools for the new lexicon file format v Lexicon file support for custom pronunciations
18
The command-line tool, voiceTest, also allows you to create a lexicon file using the -lexicon tag when defining a grammar. Both the Voice Toolkit and the voiceTest tool allow you to build lexicon files that can improve grammars. To get the command line parameter information for voiceTest, issue the following: Note: Where the WVS_ROOT variable is listed, substitute the complete path for the installation directory.
cd /$WVS_ROOT/lib ./voiceTest.sh
cd /$WVS_ROOT/lib ./voiceTest.sh
cd %WVS_ROOT%\lib voiceTest.bat
For more information on the tools provided in WebSphere Voice Toolkit, see the WebSphere Voice Toolkit Information Center.
Testing Japanese grammars

After testing your Japanese grammar using the Voice Toolkit, you might find that the WebSphere Voice Server identifies certain Japanese words as missing and produces a grammar compilation error. Use the following instructions to correct this error: 1. Open your grammar file using WebSphere Voice Toolkit. 2. Generate a lexicon file from the grammar file. v If you are using Voice Toolkit version 5.1, generate a pool file and then convert the generated .pbs file to a lexicon file (.lxml). v If you are using Voice Toolkit version 6.0.x, use the Pronunciation Builder to generate the lexicon file (.lxml). 3. Add a <lexicon> tag to the grammar file to reference the generated lexicon file. 4. Deploy the lexicon file on the Web sever with the grammar. Note: If the lexicon file is large, you should preload the grammar. For instructions on preloading grammars refer to the Configuring ASR settings section in this information center.
Tuning speech resources for effective applications

Tuning a speech application involves modifying the pronunciations of the WebSphere Voice Server automatic speech recognition (ASR) and text-to-speech (TTS) engines to recognize user input and pronounce system prompts more effectively. This section includes tables showing the phonetic pronunciation (sounds-like spelling) of vowels and consonants in each of the supported ASR languages.
19
For the TTS languages, use the Symbolic Phonetic Representation (SPR) tables listed in Using IBM Text-to-Speech Technology and Speech Synthesis Markup Language. This guide is also provided as a related document with the WebSphere Voice Toolkit Information Center. WebSphere Voice Server run-time pronunciation errors are sometimes caused by the stress placed on the vowel sounds. If you experience pronunciation errors on certain words, modify the baseforms, to improve ASR. See Preparing a lexicon of ASR pronunciations for instructions on how to do this. Use the tables in this section to assist you in correcting baseform phonetic sounds.
Preparing a lexicon of ASR pronunciations

A lexicon file enables you to define one or more pronunciations for words that will override the pronunciations provided by the Voice Server. This may be necessary for rare acronyms and foreign words that are incorrect or missing from the base pronunciation dictionary provided with the Voice Server. You can create a lexicon using an ASCII text editor of your choice, or use the voiceTest tool to generate one for the words in a grammar that are missing from the base dictionary. These words will have automatically constructed pronunciations that may not be correct. The following is a sample of an ASR lexicon where the first entry is a word with two pronunciations, the second a word with two different spellings, and the third an acronym whose pronunciation is defined as a sequence of simple words:
<?xml version=1.0 encoding=utf-8?> <lexicon version=1.0 xml:lang=en-US alphabet=x-ibmasr case-sensitive=false> <import uri=sourcelexicon.xml/> <lexeme> <spelling>preferred</spelling> <phoneme>P R AX F ER DD</phoneme> <phoneme>P R IX F ER DD</phoneme> </lexeme> <lexeme> <spelling>colour</spelling> <spelling>color</spelling> <phoneme>K AH L AXR</phoneme> </lexeme> <lexeme> <spelling>IEEE</spelling> <sounds-like>I triple E</sounds-like> </lexeme> </lexicon>
The name of the pronunciation lexicon document must be declared in the grammar source document. See Speech Recognition Grammar Specification (SRGS) 1.0 http://www.w3.org/TR/speech-grammar/ See Tuning speech resources for effective applications for details of the character symbols used to specify phonetic pronunciations and the sounds they represent in each supported language.
Canadian French phonetic pronunciation

The following table is a Canadian French phonetic chart to assist you in correcting Canadian French baseform phonetic sounds.
20
Table 1. Guide to the Canadian French phonetic alphabet Symbol | (pipe symbol) X D$ _B. _CH _D. _DF _F. _G. _H. _JJ _K. _L. _M. _N. _NG _P. _R. _RR _S. _SH _T. _T2 _TH _V. _W. _Y. _YU _Z. _Z1 _ZH A_N A_S A6S Description Word Boundary Phone The silence phone at the end of each word. The silence phone at the beginning and end of a sentence. b as in bois ch as in the English word church d as in doigt th as in the English word this f as in foire g as in gateau h as in the English word house g as in the English word page c as in coin l as in loin m as in mer n as in neuf ng as in the English word song p as in poids or p as in the English word map English r as in three French r as in noir s as in signe or s as in the English word sun ch as in chien or sh as in the English word shade t as in trois tt as in Ottowa th as in the English word thick v as in vigne or v as in the English word vote o as in loin or w as in the English word wet i as in tiers or y as in the English word you u as in huit z as in zro tz as in Fitzgerald j as in jouet or s as in the English word pleasure an as in sans a as in repas er as in the English word water
21
Table 1. Guide to the Canadian French phonetic alphabet (continued) Symbol AAL AES AVS AWD AXS AYD E_A E_N E_S E3L E9S EAA EUN EUS EYD I_L I_S ILS O_A O_N O_S OAA OAL OWD OYD U_L U_S ULS Y_S Description a as in the English word star a as in the English word back u as in the English word plum ou as in the English word out e as in ceci or e as in the English word novel i as in the English word five e as in pied, in final syllables only in as in cinq as in tude, in all non-final syllables or as in the English word word eu as in neuf e as in sept or e as in the English word devil un as in un eu as in deux ei as in the English word eight ea as in the English word sea i as in ici or i as in the English word traffic i as in sinn or i as in the English word zip o as in mot, in final syllables only on as in donc o as in roman, in all non-final syllables o as in sol, in final syllables only a as in the English word wall o as in the English word old oi as in the English word oil o as in the English word who ou as in cours u as in the English word push u as in mur
German phonetic pronunciation

The following table is a German phonetic chart to assist you in correcting German baseform phonetic sounds.
Table 2. Guide to the German phonetic alphabet Symbol | (pipe symbol) A6S Description Wortgrenzen-Phon schwaches a vi_er_, bess_er_ As in
22
Table 2. Guide to the German phonetic alphabet (continued) Symbol AXS A_N A_S A_S A_S EAS EAS EAS EUS EUS EUS E_S E_S E_S I_S I_S I_S O_N O_S O_S O_S U_S U_S U_S Y_S Y_S Y_S A_S U_S A_S I_S O_S I_S _?? _B. _C2 _CH _D. _F. _G. _H. _JJ Description obligatorisches Schwa nasaliertes a (kurz oder lang) kurzes a langes a kurzes offenes langes offenes kurzes offenes oder geschlossenes langes geschlossenes kurzes geschlossenes e langes geschlossenes e kurzes geschlossenes i langes geschlossenes i nasaliertes o kurzes offenes geschlossenes oder unsilbisches o langes geschlossenes o kurzes geschlossenes oder unsilbisches u langes geschlossenes u kurzes offenes geschlossenes oder unsilbisches langes geschlossenes au-Diphthong ei-Diphthong eu-Diphthong silbeninitialer Knacklaut b-Laut (bilabialer Plosiv stimmhaft) ich-Laut (palataler Reibelaut stimmlos) tsch-Laut (postalveolare Affrikate stimmlos) d-Laut (alveolarer Plosiv stimmhaft) f-Laut (labiodentaler Reibelaut stimmlos) g-Laut (velarer Plosiv stimmhaft) h-Laut (glottaler Reibelaut stimmlos) dsch-Laut (postalveolare Affrikate stimmhaft) As in sieb_e_n, sein_e_ or_an_ge, Ch_an_son h_a_t, _a_cht W_ah_l, B_a_d h__tte, _E_nde Ger__t, __ra V__lker, __ffnung b__se, sch__n M_e_thode, _e_gal z_eh_n, S_ee_ m_i_t, _i_n v_ie_l, Part_ie_ Sais_on,_ Pavill_on_ v_o_n, _o_ft l_o_s, B_oo_t B_u_s, _u_nd F_u_, g_u_t K__ste, B__ro m__de, gr__n _au_s, k_au_m m_ei_n, M_ai_ L_eu_te, M_u_se __aber, zu__erst _B_all, a_b_er ni_ch_t, Re_ch_t deu_tsch,_ Lan_dsch_aft _D_atum, Pa_dd_el _f_allen, A_ff_e _g_enau, Re_g_en _H_aus, Be_h_lter _Dsch_ungel, Mana_g_er
23
Table 2. Guide to the German phonetic alphabet (continued) Symbol _K. _KH _KS _L. _M. _N. _NG _P. _PF _R. _S. _SH _T. _V. _Y. _Z. _Z1 Description As in
k-Laut (velarer Plosiv stimmlos) _K_ost, La_ck_ ach-Laut (velarer Reibelaut stimmlos) ks-Laut (alveolare Affrikate stimmlos) l-Laut (alveolarer Lateral) m-Laut (bilabialer Nasal) n-Laut (alveolarer Nasal) ng-Laut (velarer Nasal) p-Laut (bilabialer Plosiv stimmlos) pf-Laut (labiodentale Affrikate stimmlos) r-Laut (Vibrant) s-Laut (alveolarer Reibelaut stimmlos) sch-Laut (postalveolarer Reibelaut stimmlos) t-Laut (alveolarer Plosiv stimmlos) Bu_ch_, do_ch_ e_x_trem, Pra_x_is _L_and, vo_ll_ _M_onat, i_mm_er _N_atur, Ha_n_d la_ng_, E_ng_e _P_ost, Schu_pp_en Ko_pf_, _Pf_anne _r_und, P_r_eis ein_s_, Wa_ss_er fal_sch,_ _S_port _T_ag, Zei_t_
w-Laut (labiodentaler Reibelaut _V_ase, _W_etter stimmhaft) j-Laut (palataler Reibelaut stimmhaft)/unsilbisches i s-Laut (alveolarer Reibelaut stimmhaft) tz-Laut (alveolare Affrikate stimmlos) _j_a, Reg_i_on ro_s_a, _s_o _z_ehn, Ka_tz_e
Japanese phonetic pronunciation

The following table is a Japanese phonetic chart to assist you in correcting Japanese baseform phonetic sounds.
Table 3. Guide to the Japanese phonetic alphabet Symbol Qb Qch Qd Qf Qg Qh Qj Qk Qky Description sokuon+b sokuon+ch sokuon+d sokuon+f sokuon+g sokuon+h sokuon+j sokuon+k sokuon+ky
24
Table 3. Guide to the Japanese phonetic alphabet (continued) Symbol Qp Qpy Qs Qsh Qt Qts Qz _n a a: b by ch d dy e e: f g gy h hy i i: j k ky m my n ny o o: p py r ry s sh t Description sokuon+p sokuon+py sokuon+s sokuon+sh sokuon+t sokuon+ts sokuon+z hatsuon vowel long vowel ba gyou youon cha gyou da gyou dya gyou vowel long vowel fa gyou ga gyou gya gyou ha gyou hya gyou vowel long vowel ji ka gyou kya gyou nasal m mya gyou nasal n nya gyou vowel long vowel pa gyou pya gyou ra gyou rya gyou sa gyou sha gyou ta gyou
25
Table 3. Guide to the Japanese phonetic alphabet (continued) Symbol ts u u: w y z Description tsu vowel long vowel wa youon za gyou
Latin American Spanish phonetic pronunciation

The following table is a Latin American Spanish phonetic chart to assist you in correcting Latin American Spanish baseform phonetic sounds.
Table 4. Guide to the Latin American Spanish phonetic alphabet Symbol AA AO AS BF BH BO CH DF DH DJ DO EA EO ES FH GF GH GO GN GS IA IJ IO IS JH KH LH LO Description casar casa casa casar cuba bosque obsoleto racha dedo dedo hoyo edad reloj jefe jefe feria liga garra signo via extremo ligar medio casi virus jarra con lado normal
26
Table 4. Guide to the Latin American Spanish phonetic alphabet (continued) Symbol LL MH NF NG NH NO OA OO OS PH RH RO RR SH SO TH UO US WH ZH ZS Description billete mar infame cinco nueve con lograr voto voto peso oro mar tierra reloj signo mes tres lugar muro cuatro cinco mismo
Simplified Chinese phonetic pronunciation

The following table is a Simplified Chinese phonetic chart to assist you in correcting Simplified Chinese baseform phonetic sounds.
Table 5. Guide to the Simplified Chinese phonetic alphabet Symbol | (pipe symbol) X D$ B Description Word boundary phone. The silence phone at the end of each word. The silence phone at the beginning and end of a sentence. b as in
BI
bi as in
BU
bu as in
27
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol C Description c as in
CU
cu as in
CH
ch as in
CHU
chu as in
d as in
DI
di as in
DU
du as in
f as in
g as in
GS
Zero initial phone as in
GU
gu as in
h as in
HU
hu as in
JI
ji as in
JU
ju as in
k as in
28
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol KU Description ku as in
l as in
LI
li as in
LU
lu as in
LYU
l as in
m as in
MI
mi as in
MU
mu as in
n as in
NI
ni as in
NU
nu as in
NYU
n as in
p as in
PI
pi as in
QI
qi as in
QU
qu as in
29
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol R Description r as in
RU
ru as in
s as in
SU
su as in
SH
sh as in
SHU
shu as in
t as in
TI
ti as in
TU
tu as in
w as in
XI
xi as in
XU
xu as in
y as in
YU
yu as in
z as in
ZU
zu as in
30
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol ZH Description zh as in
ZHU
zhu as in
A1
a1 as in
A2
a2 as in
A3
a3 as in
A4
a4 as in
A5
a5 as in
AI1
ai1 as in
AI2
ai2 as in
AI3
ai3 as in
AI4
ai4 as in
AI5
ai5 as the last vowel in
no tone. AN1 an1 as in
AN2
an2 as in
AN3
an3 as in
AN4
an4 as in
31
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol AN5 Description an5 as the last vowel in
no tone. AG1 ag1 as in
AG2
ag2 as in
AG3
ag3 as in
AG4
ag4 as in
AG5
ag5 as the last vowel in
no tone. AO1 ao1 as in
AO2
ao2 as in
AO3
ao3 as in
AO4
ao4 as in
AO5
ao5 as in
no tone. E1 e1 as in
E2
e2 as in
E3
e3 as in
E4
e4 as in
32
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol E5 Description e5 as in
no tone. EI1 ei1 as in
EI2
ei2 as in
EI3
ei3 as in
EI4
ei4 as in
EI5
ei5 as in
no tone. EN1 en1 as in
EN2
en2 as in
EN3
en3 as in
EN4
en4 as in
EN5
en5 as in
no tone. EG1 eg1 as in
EG2
eg2 as in
EG3
eg3 as in
EG4
eg4 as in
33
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol EG5 Description eg5 as in
no tone. ER1 ER2 er1 (no example available) er2 as in
ER3
er3 as in
ER4
er4 as in
ER5
er5 as the last vowel in
no tone. I1 i1 as in
I2
i2 as in
I3
i3 as in
I4
i4 as in
I5
i5 as in
no tone. IE1 ie1 as in
IE2
ie2 as in
IE3
ie3 as in
IE4
ie4 as in
34
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol IE5 Description ie5 as in
no tone. IH1 ih1 as in
IH2
ih2 as in
IH3
ih3 as in
IH4
ih4 as in
IH5
ih5 as the last vowel in
no tone. IN1 in1 as in
IN2
in2 as in
IN3
in3 as in
IN4
in4 as in
IN5
in5 as the last vowel in
no tone. IG1 ig1 as in
IG2
ig2 as in
IG3
ig3 as in
IG4
ig4 as in
35
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol IG5 Description ig5 as the last vowel in
no tone. O1 o1 as in
O2
o2 as in
O3
o3 as in
O4
o4 as in
O5
o5 as in
no tone. OG1 og1 as in
OG2
og2 as in
OG3
og3 as in
OG4
og4 as in
OG5
og5 as the last vowel in
no tone. OU1 ou1 as in
OU2
ou2 as in
OU3
ou3 as in
OU4
ou4 as in
36
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol OU5 Description ou5 as in
no tone. U1 u1 as in
U2
u2 as in
U3
u3 as in
U4
u4 as in
U5
u5 as the last vowel in
no tone. UN1 un1 as in
UN2
un2 as in
UN3
un3 as in
UN4
un4 as in
UN5
un5 as the last vowel in
no tone. YU1 yu1 as in
YU2
yu2 as in
YU3
yu3 as in
YU4
yu4 as in
37
Table 5. Guide to the Simplified Chinese phonetic alphabet (continued) Symbol YU5 Description yu5 as he last vowel in
no tone.
UK English phonetic pronunciation

The following table is a UK English phonetic chart to assist you in correcting UK English baseform phonetic sounds. This chart may also be used for Australian English because it uses the UK baseform phonetic sounds.
Table 6. Guide to the UK English phonetic alphabet Symbol | (pipe symbol) AAS AAS AES AVS OAS OAS A_S ULS AXS A_S ILS _B. _CH _D. _DF D$ E_S E3S E3S E_S ILS _F. _G. _H. ILS I_S I_S _JJ _K. _L. _M. _N. _NG AQS AXS ULS Description Word Boundary Phone As in ARE fAther cAR As in At hAm trAp As in Up bUt sOn As in Or fOUR Also As in OUt cOUnt hOW As in tablE compAny commA As in EYE fIve lIE As in But laBel caB As in CHurch naTure reaCH As in Dog meDal laD As in THe boTHer baTHe Noisy silence, such as a keystroke or door slam As in Any sAId bErry As in EArly wORd hER As in AId mAde grEY As in First telePHone rouGH As in Grab aGo biG As in Home Hat beHind As in Ink hYmn beIng As in EEl pEEp heavY As in Jay adJust chanGE As in Cat baCKer maKe As in Light vaLLey feeL As in Mat haMMer hoMe As in Night aNt baN As in siNG beiNG loNG As in hOt wAs Of As in gO knOW cOal
38
Table 6. Guide to the UK English phonetic alphabet (continued) Symbol OAS ILS _P. _R. _S. _SH _T. _TH ULS U_S U_S _V. _W. X _Y. _Z. _ZH Description As in OIl avOId toy As in Park suPPle gaP As in Right fRom thRee As in Sit miSt maSS As in SHarp naTIonal fiSH As in To aTTend paT As in Thing auThor paTH As in gOOd shOUld pUt As in OOze bOOt whO As in Very oVer saVe As in White Was tWenty Silence, such as a pause As in Yet Use bEAUty As in Zip siZe waS As in treaSure uSually viSion
US English phonetic pronunciation

The following table is a US English phonetic chart to assist you in correcting US English baseform phonetic sounds.
Table 7. Guide to the US English phonetic alphabet: vowels Sound ae ah Symbol AE AA As in add cat pan Are drop content Suggested ae ah Comments The short a sound. Rarely wrong but sometimes an open o sound is used instead of the ah. So spell mot as maht to change it from moat to mott.
uh
AH
One come
Uh or double the Very rarely wrong; following letter sometimes an unstressed sound is used instead. So spell Merom as Meromm. or Very rarely wrong. Try spelling Audi as Ordi. If ow is mispronounced as oh try switching to au or ou. The unstressed eh sound as in moment or enable but not as in enter.
Short o
AO
All off caution How about
au
AW
Ou au ow
eh
AX
enable moment
eh
39
Table 7. Guide to the US English phonetic alphabet: vowels (continued) Sound er Symbol AXR As in Suggested Comments The unstressed er sound as in another over or or but not as in her . Rare but if Byron is mispronounced more like Birron then try spelling it as Bairon or Byeron or even Byre-on. Rare but replacing the vowel letter or letters by eh usually works. For example if pleasant were mispronounced as pleesant then plehsant corrects it. Never observed but er should work. If Wacome is mispronounced Wah-come then one of Waicome Waycome or Weycome should fix. The short I sound as in it if or pick but not as in lasting. Piro should be pronounced with a short I and is mispronounced as pyro then pihro should work. The unstressed short i sound as in discuss budgeted decided or saving. If Matrio was mispronounced as mah-try-oh then matreeo matreeoh or mahtreeoh should fix. bow should be right for Beau and Dohl should work for Dole. The long i sound (or oi sound) as in avoid employ dive or why.
another over er or
Long i
AY
Fire why eye
Y ai add an e
eh
EH
Enter pleasant
eh
er
ER
Bird turn her Train eight A
er
Long a
EY
Ai ay ey
Short i
IH
If it pick
ih
ix
IX
discuss saving budgeted decided Any he obvious
ix
Long e
IY
Ee ey ie
Long o
OW
Quote open go
Oh ow oe
oi
OY
Avoid enjoy oy dive why
40
Table 7. Guide to the US English phonetic alphabet: vowels (continued) Sound Short u Symbol UH As in Good full could put Suggested uh Comments The short u sound as in good put full or could. If Clough should be pronounced as in cluf then cluhf should work. The stressed long u sound as in to use you or view. Try Bloomehn for the German pronunciation of Blumen.
Long u
UW
tooth blue view use you
Oo yoo
Usually nonvowel sounds are not a problem for the Run Time baseform engine unless the name is of foreign extraction, as in the British pronunciation of Schedule (Schxhule). The table below lists the consonant sounds if you need to select a different sound than that produced by the automatic pronunciation rules.
Table 8. Guide to the US English phonetic alphabet: consonants Sound b bd ch As in be able sob tab blurb cheap child Comments The b sound as in be or able. The hard b sound at the end of a word. The soft c sound before h. The ch sound as in cheap. The D sound as in David or and. The D sound at the end of a word. the either The voiced th sound as in the or either but NOT as in thesis or thing. The very short T or D sound as in butter greater reagarding or order. A dummy sound that marks the beginning and end of a sentence. father rough again peg log tug here who jeep roger because dark scale The soft f sound. The hard g sound. The G sound at the end of a word. The h sound. The soft g sound. The k sound but not the J sound at the beginning of a word. The k sound at the end of a word. The L sound.
d dd dh
and David
dx
butter greater regarding order
d$
f g gd hh jh k
kd L
cake level parallel
41
Table 8. Guide to the US English phonetic alphabet: consonants (continued) Sound m n As in am must final not none Comments The m sound. The starting N or the unaccented N in the middle of a word as in final not or none. The n followed by a g or k sound. The p sound but not at the beginning of a word. The p sound at the end of a word. The r sound followed by a vowel sound in the same syllable. The s not followed by an h sound. The sh sound. The t (not th) sound not at the beginning of a word. The t sound at the end of word. The unvoiced th sound but NOT as in the or either. The ts sound at the end of a word. eleven improve very frequent way question anywhere emulate yes senior you pans goes zero Asia pleasure Zho The v sound. The w sound. The y sound that leads into a vowel. The z sound. The soft z sound.
ng p
bang think singing adoption amps rapid
pd r
pop tarp mop abroad brace read
s sh t
sit circus decide action shade splash adapter retry let
td th
date quit thesis thing
ts v w y z zh
42
Chapter 3. Planning
This section explains what you need to consider when you plan a speech system using WebSphere Voice Server. The requirements of the voice server system will vary depending on the following issues: v The number of telephone lines or channels supported v v v v The call characteristics The speech application characteristics The language/voice installed The size and complexity of the grammars used (automatic speech recognition (ASR) only)
Planning your WebSphere Voice Server system involves the following steps: 1. Understanding the speech application. 2. Estimating speech resources for ASR and text-to-speech (TTS) engines. 3. Estimating resources for a LAN. 4. Identifying hardware and software requirements. 5. Creating a system topology: Single machine or multiple machines.
Understanding the speech application environment

In planning a voice system using WebSphere Voice Server speech technologies, you must determine: v The number of gateway systems you require to connect to WebSphere Voice Server. v The number of automatic speech recognition (ASR) engines and how many will be active at any one time. v The number of concatenative text-to-speech (TTS) engines, and how many will be active at any one time. v The number of machines you require, as a function of the number and speed of the processors available in each machine. v The type of local area network necessary. Note: It is not possible to give definitive information about exactly what size or number of machines you will need for your WebSphere Voice Server system in this documentation. Only approximate guidelines can be provided, and it is essential that any implementation be tested with realistic call volumes before it is put into production. For guidance about capacity planning for your specific configuration, contact your IBM representative.
Application load
To estimate the application load on the system, you need to know the following: v The number of telephony channels required to handle the number of calls you expect for applications using WebSphere Voice Server technology. You must consider how long on average each caller is expected to wait before being answered.
43
v The call frequency and distribution whether the number of calls likely to be made is the same throughout the day, or if the call pattern is likely to vary significantly, with peaks and troughs in demand. Your WebSphere Voice Server system should be able to handle the maximum demand for speech resources. That is, the resources needed at the peak calling hour rather than a days average number of hourly calls. The primary speech resource is the ASR or TTS engine. The demand for engines is influenced by both the frequency of calls and how they are distributed. If all the incoming calls use the same application and start at the same time, each call will need an engine at the same time so the demand will be high. If, on the other hand, calls are distributed normally, the number of engines needed simultaneously can be considerably smaller. For your applications, you must determine the acceptable performance or desirability of an engine being available for a call without a significant delay. Delays can cause performance degradation, such as not recognizing speech input or stuttering output. If a degradation of performance is acceptable during peak utilization, fewer engines will be required. v The time that engines are allocated, or assigned, as a proportion of a call. This is known as the allocation duty cycle. ASR and TTS engines are dynamically allocated on an as needed basis, which means they are allocated to perform a single operation (recognition or synthesis) and then immediately freed. v The time that engines are active (recognizing speech or synthesizing text to be played) as a proportion of a call. This is known as the active duty cycle. This can vary considerably, depending on the design of your voice applications and their complexity. These factors determine the number of concurrent ASR and TTS sessions that are required. The number of concurrent ASR and TTS sessions, in turn, determines the number of processors required and how powerful they must be. Similarly, these two variables number and speed of the processors dictate the number and size of the machines needed for your WebSphere Voice Server installation. For example, a non-barge-in application using long prompts of synthesized text together with a simple grammar is likely to be actively engaged in recognition for only a short proportion of the length of a call. It will have a short active duty cycle for ASR but a long active duty cycle for speech synthesis. A barge-in application using shorter prompts of synthesized text together with a complex grammar is likely to spend more time actively engaged in recognition. In this case, the active duty cycle for ASR will be longer and the active duty cycles for speech synthesis shorter. If a system is underspecified, an engine might not be available at the start of a call. The following figures show a comparison of ASR and TTS engine allocation. In this figure, a barge-in application is being used.
44
Figure 2. ASR and TTS engine allocation comparison (speech recognition with barge-in)
In this figure, a non-barge-in application is being used.
Chapter 3. Planning
45
Figure 3. ASR and TTS engine allocation comparison (speech recognition without barge-in)
Application design
Some applications can be designed so that most callers are likely to complete their business and hang up quickly, thus freeing resources for another call. For more information about application design, refer to the WebSphere Voice Toolkit Information Center.
System resource
Once you know the maximum number of concurrent sessions required for ASR and TTS, the recognition languages and voices to be used, and the complexity of your voice recognition applications, you can determine how many WebSphere Voice Server machines are necessary. You can also determine the minimum specifications for the machines, which are also dependent on the operating system you select. Remember: WebSphere Voice Server must run on its own machine. A minimum base memory of 2 GB is required on each WebSphere Voice Server recognition server machine. You can minimize the number of machines required by installing multiple high-speed processors and additional memory in each machine. The actual number of WebSphere Voice Server engines that will run on each machine is solution-dependent. A solution must be tested to verify that a system can handle a condition where all of the WebSphere Voice Server engines are fully utilized. It also is important to ensure that the qualified compatible gateway system can support all of the attached server machines and engines.
46
Related information Estimating resources for ASR engines Use this section to understand the resources required for automatic speech recognition (ASR) engines. Estimating resources for TTS engines Use this section to understand the resources required for text-to-speech (TTS) engines.
Estimating resources for ASR engines

Use this section to understand the resources required for automatic speech recognition (ASR) engines. The critical resources are CPU and memory. CPU is consumed when an ASR engine is processing speech. Memory is consumed when ASR engines are allocated or grammars loaded. See Tuning WebSphere Voice Server on page 145 for a rule of thumb approach to engine allocation. CPU utilization is determined by: v The active duty cycle of the started ASR engines (application load) v The size and complexity of the grammars required by your voice applications v The language being recognized v Use of barge-in Memory utilization is determined by: v The number of ASR engines allocated v The languages installed v The size and complexity of the grammars required by your voice applications Note: There are limits to the CPU load that is desirable. If the CPU utilization is too high, application response time can increase.
Estimating resources for TTS engines

Use this section to understand the resources required for text-to-speech (TTS) engines. The critical resources are CPU and memory. CPU is consumed when a TTS engine is synthesizing text. Memory is consumed when TTS engines are allocated. See Tuning WebSphere Voice Server on page 145 for a rule of thumb approach to engine allocation. CPU utilization is determined by: v The active duty cycle of the started TTS engines (application load) v The language or voice synthesized from text v The number of words or utterances in a singles instance of text to be synthesized v The ability of the application to take advantage of caching Memory utilization is determined by: v The number of TTS engines allocated v The languages or voices installed and available for use
Chapter 3. Planning
47
v The size of the cache (default is 20 MB) Note: The fifth generation voices require more memory and CPU than the other voices. Depending on the capacity of your WebSphere Voice Server, you might need to do the following when using these voices: v Reduce the number of channels supported v Increase the memory or CPU During initialization of WebSphere Voice Server, all available voices for the installed languages are loaded into a shared memory for use by any TTS engine. The voices remain in memory until the WebSphere Voice Server is stopped. In addition, an output cache is created in shared memory to hold recently synthesized phrases. The use of an output cache can improve synthesis processing, depending on the design of the application and whether the same phrase is used repeatedly. Also refer to Configuring default and available TTS voices on page 104 and TTS troubleshooting tips on page 463 for tips on optimizing the performance of your WebSphere Voice Server.
Estimating resources for a LAN

To handle the transport of voice data and control information between the client and the server, connect the WebSphere Voice Server to the qualified compatible VoiceXML browser and gateway using a dedicated local area network (LAN). The minimum requirement is for a 100 Mbps Ethernet network, or equivalent. The LAN technology used must be appropriate for the number of channels and by extension, the number of automatic speech recognition (ASR) and text-to-speech (TTS) engines, configured for your installation. It is difficult to be precise in estimating the saturation point for a specific configuration. If you have questions about these issues, contact your IBM representative who can give you advice on capacity planning.
Identifying hardware and software requirements

Hardware requirements
Supported hardware for WebSphere Voice Server includes machines that meet the minimum hardware criteria defined below.
Table 9. Minimum supported hardware Processor Memory (RAM) 2 GB (3 GB if you plan to install fifth generation voices) Free Disk Space Network 2 GB (3 GB if you plan to install fifth generation voices) TCP/IP Other CD
Intel x86 1 GHz or equivalent processor pSeries at 450 MHz
48
Additionally for AIX, make sure you provide adequate disk space for the following directories: v 450 MB for the /usr/WebSphere/AppServer directory The installation root directory includes the base WebSphere Application Server product code. v 600 MB for the /tmp directory The temporary directory is the working directory for the installation program. v 20 MB for the /usr/IBMHttpServer directory The IBM HTTP Server requires this space. v 35 MB for the /usr/IBM/gsk7 directory The Tivoli Global Security Kit requires this space.
Table 10. Text-to-speech (TTS) server disk space requirements by voice Language WebSphere Voice Server Male French Canadian German Japanese Latin American Spanish Simplified Chinese UK English US English 177 MB 287 MB 63 MB 184 MB Female/2nd 292 MB 339 MB 322 MB 349 MB Fifth generation
n/a 286 MB 275 MB
628 MB 265 MB 317/321 MB Kate - 1.3 GB (600 MB RAM) Julie - 1.6 GB (600 MB RAM) Lisa - 1.5 GB (600 MB RAM)
Software requirements
WebSphere Voice Server supports the following operating systems: v v AIX 5L version 5.3 with maintenance level (ML) 3 Red Hat Enterprise Linux WS/ES/AS for Intel 3.0 with Update 1, Update 3, or Update 4 (2.4 Kernel) - supported in WebSphere Voice Server version 5.1.3
Red Hat Enterprise Linux WS/ES/AS for Intel 3.0 with Update 1 only (2.4 Kernel) - supported in WebSphere Voice Server versions 5.1.1 and 5.1.2 v SUSE LINUX Enterprise Server 8, powered by United Linux 1.0 (SLES-8) (Intel) with Service Pack 2a or 3 v v Windows Server 2003 Enterprise or Standard editions with or without Service Pack 1 Note: If you are using Service Pack 1, download and apply the WebSphere Voice Server interim fix, IC47845, from the product Support site.
Chapter 3. Planning
49
See the WebSphere Voice Server system requirements page for the latest information about supported operating systems.
Compatibility with IVRs and gateways

For a listing of the current compatible Interactive Voice Response (IVR) systems and gateways for WebSphere Voice Server, see IVR and gateway compatibility for WebSphere Voice Server for Multiplatforms on the Web.
Single machine topology

The maximum number of concurrent sessions a single machine can support is heavily dependent upon your hardware, the number and speed of the processors, and the amount of RAM on the machine. For availability and failover support, you might want to use multiple machines. With a single machine, if you take the server down for maintenance, your production environment is down. A single machine can be managed through the WebSphere Application Server administrative console providing ease of use for configuration, monitoring, tracing, alarming and logging. The WebSphere Voice Server with the WebSphere Application Server base product runs on one machine. You can install the product in a standalone configuration or as part of a Load Balancer cluster in a multiple machine configuration. The standalone configuration is typically for developer desktops or standalone production computing, which involve a single Application Server instance operating independently of any other applications.
Different single machine topologies
50
Figure 4. The telephony environment
WebSphere Voice Server supports Media Resource Control Protocol (MRCP) acting as an MRCP Server. MRCP is defined in the Internet Engineering Task Force (IETF). A VoiceXML browser that supports MRCP is considered an MRCP Client. A qualified VoiceXML browser and gateway allows for telephony access to WebSphere Voice Server. If the VoiceXML browser runs inside a gateway, an HTTP Server might be required to provide the VoiceXML documents, audio, and grammar files because the gateway might be constrained regarding memory.
Chapter 3. Planning
51
Figure 5. The desktop environment
For development purposes, a limited set of functionality is provided to test WebSphere Voice Server without requiring a VoiceXML browser, gateway, and telephony infrastructure. This functionality is provided by the voiceTest program acting as an MRCP Client. The voiceTest program enables: v Testing a grammar against a prerecorded audio utterance and receiving the recognition result. v Synthesizing a text document and saving the synthesized speech to an audio file. The audio file can be played to hear the results. v Enhancing the lexicon documents referenced by a grammar. When using the voiceTest program as the MRCP client, access to local audio and grammar files is provided by the file URI. An optional HTTP server can be used to provide the audio and grammar files, however, it must be installed on a separate machine (see the preceding Voice Test environment figure). For more information on command line parameter information for voiceTest refer to the Testing grammars section in this information center. For more information on the tools provided in WebSphere Voice Toolkit, see the WebSphere Voice Toolkit Information Center.
52
Multiple machine topology

Multiple machine environments extend basic single-machine WebSphere Voice Server configurations by distributing the voice server solution over multiple machines, increasing the overall processing power from one machine to contributions from all machines in the configuration. Multiple machine environments use two other WebSphere Application Server products included in the WebSphere Voice Server packaging: v Network Deployment Allows a single WebSphere Application Server administrative console to manage a cell of application server machines running either WebSphere Voice Server or customer applications. v Load Balancer Allows for scalability, throughput, availability, failover, and maintainability.
Figure 6. Network Deployment topology
The preceding figure shows Network Deployment managing a cell of voice servers with different languages installed. An HTTP server is required to provide audio and grammar files to WebSphere Voice Server. Refer to the Administering section for more information.
Advantages of multiple machine topologies

You can configure multiple machines to add processing power, improve security, maximize availability, and balance workloads. Network Deployment and Edge Components provide clusters, workload management, and the Load Balancer to
Chapter 3. Planning
53
implement configurations that address these issues. These scaling techniques are generally combined to maximize benefits and minimize problems associated with multiple machine systems. Topologies that incorporate more than one cell also have the following advantages: v Isolation of hardware failure. If one cell goes offline due to hardware problems, the others can still process client requests. v Isolation of software failure. Running different voice servers in two or more cells isolates any problems that occur within a cell, while the other cells continue to handle client requests. This isolation is helpful in a variety of situations: v When rolling out a new voice server or a revision of an existing voice server. You can bring the new voice server or revision online in one cell, and test it in a live situation while other cells continue to handle client requests with the production version of the voice server. v When deploying a new version of the WebSphere Voice Server and WebSphere Application Server software. You can bring the new version into production, and test it in a live situation without interrupting service. v When applying fixes or patches to the WebSphere Voice Server and WebSphere Application Server software. You can take each cell offline to upgrade it, without interrupting the application. If an unforeseen problem occurs with the new software, using multiple cells can prevent an outage to an entire site. You can also rollback to a previous software version more quickly. You can handle hardware and software upgrades on a cell-by-cell basis during off-peak hours.
Challenges of multiple machine topologies

Using multiple cells has a couple of challenges: v Deployment is more complicated than for a single administrative cell. v Multiple cells require more administration effort because each cell is administered independently. Use scripts to standardize and automate common administrative tasks to reduce this problem. For more information about multiple machine benefits and issues, refer to the following topic in the WebSphere Application Server information center: Multimachine topology concepts
Establishing multiple machine environments

Perform the following steps to set up a Network Deployment environment on each of the systems that comprise the cell.
Steps for this task

1. Install WebSphere Voice Server (which installs the base WebSphere Application Server product) on each machine to create a WebSphere Voice Server node on the machine. The installation procedure is the same for a node that federates into, or joins, a cell as it is for a stand-alone WebSphere Voice Server.
54
Important: Make sure you run the installation verification test (IVT) for WebSphere Voice Server before federating a server into a cell. 2. Install the WebSphere Application Server deployment manager on the system administration machine. Remember: v Only one deployment manager host machine is allowed for each cell. As the deployment manager federates, or joins, base WebSphere Application Server nodes, it expands the cell that it manages. v Unless you have a machine with the capacity to host both products, do not install a base WebSphere Application Server on the same machine as the deployment manager in a production environment. v The only functions supported in the deployment manager installation are the deployment manager and its associated administrative programs, such as the administrative console. 3. Start the deployment manager process. There are two ways to start the deployment manager server: v As a monitored process, which restarts automatically if a failure occurs. v As an unmonitored process, which is what the startManager.sh or startManager.bat script does. Depending on your operating system run the startManager.sh or startManager.bat script from the /bin directory of the installation root of the deployment manager. For production systems, running the deployment manager as a monitored process is recommended. 4. Run the addNode.sh or addNode.bat script on every node that you plan to federate into the cell. The addNode.sh and addNode.bat scripts incorporate a base WebSphere Voice Server product node into a deployment manager cell. You must run this tool on every system that you plan to make part of a Network Deployment cell. There are several parameters for use with the addNode command, but the minimum required set are the host name of the deployment manager node, the JMX connector type, the JMX port of the deployment manager node, and the includeapps flag. For example:
addNode wvsdmgr 8879 -includeapps
The example adds the base node on which the command runs to the cell managed by the wvsdmgr deployment manager node, using the default SOAP JMX connector type at port 8879. The command includes all applications on the base node into the cell. As it federates the base node in response to the addNode script, the deployment manager also instantiates the node agent process,nodeagent, on the WebSphere Voice Server node. 5. Start the newly federated WebSphere Voice Server. 6. Complete the set up of the new WebSphere Voice Server by configuring performance monitoring infrastructure (PMI) monitoring for the newly federated node. 7. If necessary, enable the appropriate level of security after the installation is complete.
What to do next
Now that you have created a WebSphere Voice Server cell, use the administrative console or other administrative tools to monitor and control the incorporated
Chapter 3. Planning
55
nodes, and the resources on these nodes. The console provides a central location for configuring, monitoring, and controlling all voice servers on all nodes within the cell.
56
Chapter 4. Installing
WebSphere Voice Server uses a graphical LaunchPad and interface wizard that will guide you through the installation process. After installation, a First Steps interface provides a convenient place to explore the product. For WebSphere Voice Server version 5.1.x, the LaunchPad and the installer wizard are in English. Installing WebSphere Voice Server includes the following steps: 1. Determine the type of installation: Single machine or multiple machines 2. Start the installation using the installation wizard or install silently. 3. Verify the installation. 4. Explore WebSphere Voice Server.
Installing WebSphere Voice Server

This topic introduces you to the tasks you need to perform when installing WebSphere Voice Server.
Before you begin

To create a complete environment for installation, you must install any prerequisites and prepare the operating platform for installation. After installation, you can run the installation verification test and bring up WebSphere Voice Server.
What you need to know

Perform the following tasks to create a running version of the product on your machine.
Steps for this task

1. Plan to install an e-business network. This task helps you plan to install an e-business network. It also describes interoperability considerations. 2. Install the product. This task helps you prepare your machine for installation and explains the different types of installation available to you. It includes information on: v Using the LaunchPad v Using the silent installation method This task describes using the installation wizard. Through the installation wizard, you perform some initial configuration. 3. Verify the installation. a. Use the firststeps tool to run the Installation Verification Test (IVT). Start the firststeps tool at the end of the installation. b. Identify and correct any problems using the troubleshooting procedure. 4. Install the most current fix pack for WebSphere Voice Server. 5. Set up a multinode environment.
57
This task describes how to set up multiple nodes into a group or cell of voice servers, with centralized configuration under the control of the deployment manager node and its distributed node agents. 6. Automatically restart WebSphere Application Server processes. This task describes how to set up WebSphere Application Server processes for the operating system to monitor and restart. By default, WebSphere Voice Server starts when WebSphere Application Server starts.
Results
The WebSphere Voice Server product is installed. See Uninstalling the product for information about uninstalling WebSphere Voice Server and its related products.
WebSphere Voice Server package

WebSphere Voice Server ships with a subset of WebSphere Application Server Network Deployment CDs and with a CD that contains an IBM TTS connector for the Genesys Voice Platform (GVP) NE 6.5.4 and EE 6.5.5. Note: All WebSphere Application Server Network Deployment CDs are included for use by WebSphere Voice Server only. The WebSphere Application Server base CD is not to be used on any machine other than the one on which WebSphere Voice Server is running. The WebSphere Application Server and included IBM HTTP Server are not to run any user applications or host user files. The included WebSphere Application Server Network Deployment CDs for all supported platforms are the following: v WebSphere Application Server CD The WebSphere Application Server product installation image contains the core application server runtime, IBM HTTP Server , and the node agent for communicating to the deployment manager when it is part of a cell. Note: The WebSphere Voice Server installation silently installs WebSphere Application Server. v Deployment Manager CD The Deployment Manager product installation image contains the deployment manager configured for use in departmental production computing scenarios. Although the Network Deployment package includes the base Application Server CD, installing the Network Deployment product does not install the base Application Server product. You must use the base product CD to install the base product. Note: Use the Deployment Manager installation on the CD after using the WebSphere Voice Server base CD to install the voice system management functionality. v Edge Components CD The Edge components installation image contains the Load Balancer for failover, recovery, serviceability, and scalability. For more information about the complete WebSphere Application Server Network Deployment package, see the WebSphere Application Server Network Deployment 5.1.x Information Center.
58
Installing the product

This topic describes how to install WebSphere Voice Server as the root user on AIX or Linux or from an administrator group ID on Windows Server 2003 Enterprise and Standard editions. The WebSphere Voice Server installation also installs WebSphere Application Server Network Deployment. This topic provides enough detail to guide you through preparing for, choosing, and installing the variety of options and features provided.
Before you begin

Read through this topic, and its related topics, to prepare for installation and to make yourself familiar with installation options, before you use the installation tools. WebSphere Voice Server supports the same operating systems as WebSphere Application Server Network Deployment. WebSphere Application Server Network Deployment 5.1.x (which includes Deployment Manager and Load Balancer) supports the following: v v AIX 5L version 5.3 with maintenance level (ML) 3 Red Hat Enterprise Linux WS/ES/AS for Intel 3.0 with Update 1, Update 3, or Update 4 (2.4 Kernel) - supported in WebSphere Voice Server version 5.1.3
Red Hat Enterprise Linux WS/ES/AS for Intel 3.0 with Update 1 only (2.4 Kernel) - supported in WebSphere Voice Server versions 5.1.1 and 5.1.2 v SUSE LINUX Enterprise Server 8, powered by United Linux 1.0 (SLES-8) (Intel) with Service Pack 2a or 3 v Windows Server 2003 Enterprise or Standard editions with or without Service Pack 1 v Note: If you are using Service Pack 1, download and apply the WebSphere Voice Server interim fix, IC47845, from the product Support site. Read these topics before installing the product: v WebSphere Voice Server system requirements page v Platform-specific tips for installing and migrating v Installation prerequisites v Using the LaunchPad to start the installation v Installing silently If you encounter a problem such as not having enough temporary space or not having prerequisite packages on your system, cancel the installation, make the required changes, and restart the installation. The LaunchPad tool lets you access the Getting Started Guide and or other related documents. After starting the LaunchPad, click the Install the product option to begin the installation wizard, which: v Automatically checks prerequisites v Asks the user for a full installation (which includes the WebSphere Application Server and the IBM HTTP Server installation) or custom installation.
59
Remember: To install or uninstall using a Linux or an AIX platform, log in as a root user. To install or uninstall using a Windows platform, log in as a user with Administrator privileges.
Installing
v Installing version 5.1.3 on page 68 v Installing version 5.1.2 and earlier versions on page 62
Prerequisites
Before installing WebSphere Voice Server, identify the hardware and software you require, and then refer to the topics below for any additional prerequisites.
Configure a static IP address for all operating systems

WebSphere Voice Server requires a static IP address. Configure your operating system network settings to use a static IP address and confirm that the hosts file is set up correctly. The hosts file is in the /etc/hosts directory on Linux or AIX. The file is in the C:\Windows\system32\drivers\etc\hosts directory on Windows. For example:
# Do not remove the following line, or various programs # that require network functionality will fail. 34.34.130.1 wvs.ibm.com wvs 127.0.0.1 localhost.localdomain localhost
Installation prerequisites for AIX

There is a port number conflict between the WebSphere Application Server administrative console and WebSM service on AIX. Both use port 9090. The port 9090 conflict on AIX is a known WebSphere Application Server issue. Ensure that no other applications are using this port, and make sure that the port 9090 is commented out of the /etc/services file. Check to see if you have the WebSM Service enabled on your machine by issuing the following command: lslpp -h sysmgt.websm.framework If the command fails, then you do not have the WebSM Service installed, but make sure that the port 9090 is commented out of the /etc/services file. If you do have the WebSM Service enabled, disable it by issuing the following command: /usr/websm/bin/wsmserver -disable
Installation prerequisites for Red Hat Enterprise Linux V3.0

WebSphere Voice Server version 5.1.3 supports Red Hat Enterprise Linux WS/ES/AS for Intel 3.0 with Update 1, Update 3, or Update 4 (2.4 Kernel). WebSphere Voice Server versions 5.1.2 and 5.1.1 support Red Hat Enterprise Linux WS/ES/AS for Intel 3.0 with Update 1 only (2.4 Kernel). A known limitation exists in the installation wizard when it examines prerequisite packages on Linux systems. If you have not completed a full Red Hat install, you must install the following packages on Red Hat Enterprise Linux V3.0 for xSeries platforms before installing WebSphere Voice Server: v compat-gcc-7.3-2.96.122 v compat-libstdc++-7.3-2.96.122
60
v v v v v
compat-libstdc++-devel-7.3-2.96.122 compat-glibc-7.x-2.2.4.32.5 compat-gcc-c++-7.3-2.96.122 compat-db-4.0.14-5 rpm-build-4.2.1-4.2
These packages can be found in the Red Hat Legacy S/W Development and Development tools packs. Also see the WebSphere Voice Server system requirements page for the latest information about supported operating systems.
Using the LaunchPad to start the installation

Use the LaunchPad to install the product and to access the product overview and the Getting Started Guide. Remember: To perform this task using a Linux or an AIX platform, log in as a root user. To perform this task using a Windows platform, log in as a user with Administrator privileges. v Steps for this task using an AIX platform v Steps for this task using a Linux platform v Steps for this task using a Windows platform
Steps for this task using an AIX platform

1. Open a terminal window and issue a fully qualified launchpad.sh command to start the LaunchPad, for example mount/cdrom/launchpad.sh. The LaunchPad program is in the AIX root directory of the product CD. Note: Mount the CD-ROM drive on AIX platforms. Consult the installation procedure for your platform for instructions on mounting CD-ROM drives, or refer to Mounting a CD-ROM drive on AIX. The LaunchPad opens in English only. 2. Use the LaunchPad to access the product overview and the Getting Started Guide. 3. Click Install the product to launch the installation wizard.
Steps for this task using a Linux platform

1. Open a terminal window and issue a fully qualified launchpad.sh command to start the LaunchPad, for example mount/cdrom/launchpad.sh. The LaunchPad program is in the Linux root directory of the product CD. Note: Mount the CD-ROM drive on Linux platforms. Consult the installation procedure for your platform for instructions on mounting CD-ROM drives. The LaunchPad opens in English only. 2. Use the LaunchPad to access the product overview and the Getting Started Guide. 3. Click Install the product to launch the installation wizard.
Steps for this task using a Windows platform

1. When you insert the product CD, the LaunchPad should start automatically.
61
Note: If it does not start automatically, open the Windows Explorer to the drive the CD is in and double-click on the Launchpad.exe file. The LaunchPad opens in English only. 2. Use the LaunchPad to access the product overview and the Getting Started Guide. 3. Click Install the product to launch the installation wizard.
Installing version 5.1.2 and earlier versions

This topic describes how to install WebSphere Voice Server 5.1.2 and earlier versions. Remember: To install or uninstall using a Linux platform, log in as a root user. To install or uninstall using a Windows platform, log in as a user with Administrator privileges. v Accessible options for installing WebSphere Voice Server v Installing on Linux v Installing fix pack 1 for WebSphere Application Server 5.1 on Linux on page 65 v Installing on Windows Server 2003 Enterprise or Standard editions on page 65 v Installing fix pack 1 for WebSphere Application Server 5.1 on Windows on page 67 v Post-installation steps on page 68 v Results on page 68
Accessible options for installing WebSphere Voice Server

The installation instructions in the following sections use the Launchpad tool. If you cannot use the Launchpad tool, you can launch the installation directly by issuing the install command: CD_mount_point/dist/install.sh CD_drive\install.bat
Installing on Linux
Tip: WebSphere Application Server and any required fix packs are installed silently with WebSphere Voice Server. While existing installations of WebSphere Application Server are supported, installing WebSphere Application Server before installing WebSphere Voice Server is not necessary. 1. Prepare your operating system platform for installation by verifying that you have adequate disk space and that you have the necessary installation prerequisites before starting installation. 2. Insert the WebSphere Voice Server for MRCP Linux CD. The LaunchPad tool automatically launches the Installation wizard, which is an InstallShield for MultiPlatforms (ISMP) application. The Installation wizard is in English. You can also start the Installation wizard from the product CD using the LaunchPad (launchpad.sh). Always use the fully qualified path to the launchpad.sh when installing the product. The following examples show how to start the launchpad.sh when installing from the product CD on a Linux system:
62
a. Place the CD in the drive. b. Issue the appropriate command for your Linux version in order to mount the CD and start the LaunchPad tool. For Red Hat:
cd / mount /mnt/cdrom /mnt/cdrom/launchpad.sh
For SUSE:
cd / mount /media/cdrom /media/cdrom/launchpad.sh
To unmount the CD, issue the following commands: For Red Hat:
cd / umount /mnt/cdrom
For SUSE:
cd / umount /media/cdrom
Failing to use the correct path can cause ISMP errors that stop the installation. Tip: You can start the installation using the install.sh command in the dist directory of the CD to bypass the LaunchPad. This is required for silent installation. The rest of this procedure assumes that you are using the Installation wizard, launched from either install.sh or LaunchPad. Corresponding entries in the response file exist for every prompt that is described as part of the wizard. Review the description of the response file for more information. Comments in the file describe how to customize its options. The LaunchPad Welcome panel displays. Click Install the product to continue. In the InstallShield Welcome panel, click Next to continue. The license agreement appears for you to read. Click the radio button beside the I accept the terms in the license agreement message if you agree to the license agreement and click Next to continue. After you accept the licensing terms, the installation wizard checks for prerequisites and for previous versions. Next, the language panel for WebSphere Voice Server displays. Select one of the languages for WebSphere Voice Server language capability, as well as for TTS and ASR capabilities.
3. 4. 5.
6.
Note: WebSphere Voice Server supports only one language installation. 7. The Setup screen prompts you to select either a Full or Custom installation. v Select the Full radio button if you are installing WebSphere Application Server and WebSphere Voice Server. Click Next to continue. v Select the Custom radio button if you are installing WebSphere Voice Server on an existing instance of WebSphere Application Server. Click Next to continue. 8. If you are installing WebSphere Application Server for the first time, you will be prompted for the following: a. Choose installation directories for WebSphere Application Server and for the IBM HTTP Server. b. Choose a node name and a host name for WebSphere Application Server.
63
9. 10. 11. 12.
If you already have WebSphere Application Server installed, you are prompted to choose to use an existing instance and click the appropriate radio button. Then click Next to continue. If an incorrect location for WebSphere Application Server is displayed, enter the correct location in the field at the bottom of the panel and click Next to continue. Configure your WebSphere Application Server as a service when prompted. Choose the destination directory for WebSphere Voice Server. Click Next to continue. A summary panel displays your installation selections. Click Next to start the installation process. You are then prompted to provide various CDs or image locations for each component.
Note: The TTS installation is lengthy. Do not cancel out of it. The Summary panel indicates when the installation is complete and displays the installation status. 13. A summary panel of the installation with instructions to reboot appears. Click Finish to close the installation wizard. 14. Prior to rebooting your system, add the wvs_install_path/wvs.setenv line to your shells startup profile (.bash_profile if you are using the bash shell on RedHat Linux). Edit the appropriate operating system level shell startup file and add the line:
. /opt/WebSphere/VoiceServer/wvs.setenv
Ensure that the startup profile runs during start up. 15. If you receive any warnings during installation, examine the /tmp/CWVOptInstall.log file to verify that there were no file system problems or other unusual errors. If there are problems, correct them, uninstall the product, and then reinstall it. 16. Restart the system and run the WebSphere Voice Server firststeps command on page 85. The command script is located at $WVS_ROOT/firststeps/firststeps.sh, where $WVS_ROOT is a variable defined during the installation of WebSphere Voice Server. Issue the following commands at a command prompt:
cd $WVS_ROOT/firststeps ./firststeps.sh
On the First Steps panel, click Start the server. Then, once you receive the Open for e-business message that indicates the server started, click Verify installation to make sure that all is well with your installation. If any failures are reported, see Installation troubleshooting tips on page 461. Remember: The Linux firewall must be disabled for the Interactive Voice Response (IVR) to communicate with WebSphere Voice Server. 17. Apply the appropriate tuning parameters. 18. Install the most current fix pack for WebSphere Voice Server. Remember: The Linux firewall must be disabled for the Interactive Voice Response (IVR) to communicate with WebSphere Voice Server.
64
Installing fix pack 1 for WebSphere Application Server 5.1 on Linux

Now, you must obtain fix pack 1 for WebSphere Application Server Version 5.1 Base and Network Deployment from the WebSphere Application Server support site and install it on your Linux systems. 1. Locate fix pack 1 and download the Linux on Intel Base package. Note: If you are applying fix pack 1 to your Deployment Manager installation, download the Linux on Intel ND package. 2. Make a temporary directory for the fix pack:
cd /tmp mkdir WAS511
3. Change directory to the new directory:

cd WAS511
4. Copy the fix pack file that you downloaded, was51_fp1_linux.tar.gz, to the new directory. 5. Unzip and untar the file:
gunzip was51_fp1_linux.tar.gz tar -xvf was51_fp1_linux.tar
6. Reboot the system and ensure that WebSphere Application Server, IBM HTTP Administration Server, and IBM HTTP Server are stopped. 7. Set up the WebSphere Application Server command line:
cd WAS_installation_directory/bin . ./setupCmdLine.sh
WAS_installation_directory is the installation directory where you installed WebSphere Application Server. When you run the setupCmdLine.sh command, the $WAS_HOME variable is defined for this location. 8. Change to the fix pack directory and start the update wizard to install the fix pack:
cd /tmp/WAS511/ ./updateWizard.sh
9. Install the fix pack by following the instructions in the wizard and selecting the defaults. 10. Reboot the system. 11. After installing fix pack 1 for WebSphere Application Server 5.1 and starting WebSphere Application Server, rerun the WebSphere Voice Server firststeps command and verify your installation again. 12. Then restore the WebSphere Voice Server Collector. Change directory to $WAS_HOME/bin and issue the restoreWVSCollector.sh command.
Installing on Windows Server 2003 Enterprise or Standard editions

Tip: WebSphere Application Server and any required fix packs are installed silently with WebSphere Voice Server. While existing installations of WebSphere Application Server are supported, installing WebSphere Application Server before installing WebSphere Voice Server is not necessary. Remember: Installing on Windows Server 2003 is only supported on WebSphere Voice Server version 5.1.2 and later.
65
Note: If you are using Service Pack 1, download and apply the WebSphere Voice Server interim fix, IC47845, from the product Support site. 1. Prepare your operating system platform for installation by verifying that you have adequate disk space and that you have the necessary installation prerequisites before starting installation. 2. Insert the WebSphere Voice Server for MRCP Windows Server 2003 CD. The LaunchPad tool automatically launches the installation wizard. If necessary, you can also start the installation wizard from the product CD by double-clicking the launchpad.exe file. This action launches the installation wizard in English. Tip: You can start the installation using the install.bat command in the base directory of the CD to bypass the LaunchPad. This is required for silent installation. The LaunchPad Welcome panel displays. Click Install the product to continue. The system displays an Install program has been launched message. Click OK. In the InstallShield Welcome panel, click Next to continue. Click the radio button beside the I accept the terms in the license agreement message if you agree to the license agreement and click Next to continue. After you accept the licensing terms, the installation wizard checks for prerequisites and for previous versions. Next, the language panel for WebSphere Voice Server displays. Select one of the languages for WebSphere Voice Server language capability, as well as for TTS and ASR capabilities.
3. 4. 5. 6.
7.
Note: WebSphere Voice Server supports only one language installation. 8. The Setup screen prompts you to select either a Full or Custom installation. v Select the Full radio button if you are installing WebSphere Application Server and WebSphere Voice Server. Click Next to continue. v Select the Custom radio button if you are installing WebSphere Voice Server on an existing instance of WebSphere Application Server. Click Next to continue. 9. If you are installing WebSphere Application Server for the first time, you will be prompted for the following: a. Choose installation directories for WebSphere Application Server and for the IBM HTTP Server. b. Configure both WebSphere Application Server and HTTP Server as a service. c. Enter a logon user ID and a password. Make sure that the logon user ID belongs to the administrator group and has the following advanced user rights: v Act as part of the operating system v Log on as a service d. Choose a node name and a host name for WebSphere Application Server. If you already have WebSphere Application Server installed, you are prompted to choose to use an existing instance and click the appropriate radio button. Then click Next to continue.
66
If an incorrect location for WebSphere Application Server is displayed, enter the correct location in the field at the bottom of the panel and click Next to continue. 10. Choose the destination directory for WebSphere Voice Server. Click Next to continue. 11. A summary panel displays your installation selections. Click Next to start the installation process. 12. You are then prompted to provide various CDs or image locations for each component. Note: The TTS installation is lengthy. Do not cancel out of it. The Summary panel indicates when the installation is complete and displays the installation status. 13. A summary panel of the installation with instructions to reboot appears. Click Finish to close the installation wizard. 14. If you receive any warnings during installation, examine the %TEMP%\CWVOptInstall.log file to verify that there were no file system problems, or other unusual errors. If there are problems, correct them, uninstall the product, and then reinstall it. 15. Restart the system and run the WebSphere Voice Server firststeps command on page 85. The command script is located at %WVS_ROOT%\firststeps\firststeps.bat, where %WVS_ROOT% is a variable defined during the installation of WebSphere Voice Server. Issue the following commands at a command prompt:
cd %WVS_ROOT%\firststeps firststeps.bat
On the First Steps panel, click Start the server. Then, once you receive the Open for e-business message that indicates the server started, click Verify installation to make sure that all is well with your installation. If any failures are reported, see Installation troubleshooting tips on page 461. 16. Apply the appropriate tuning parameters. 17. Install the most current fix pack for WebSphere Voice Server.
Installing fix pack 1 for WebSphere Application Server 5.1 on Windows

Now, you must obtain fix pack 1 for WebSphere Application Server Version 5.1 Base and Network Deployment from the WebSphere Application Server support site and install it on your Windows systems. 1. Locate fix pack 1 and download the Windows Base package. Note: If you are applying fix pack 1 to your Deployment Manager installation, download the Windows ND package. Create a temporary directory for the fix pack. Copy the fix pack file that you downloaded, was51_fp1_win.zip, to the new directory and extract its contents. Ensure that WebSphere Application Server, IBM HTTP Administration Server, and IBM HTTP Server are stopped. Set up the WebSphere Application Server command line. Change directory to WAS_installation_directory\bin and issue the setupCmdLine.bat command.
2. 3. 4. 5.
67
6. 7. 8. 9.
10.
WAS_installation_directory is the installation directory where you installed WebSphere Application Server. When you run the setupCmdLine.bat command, the %WAS_HOME% variable is defined for this location. Change to the fix pack directory and issue the updateWizard.bat command. To complete the installation, follow the instructions in the wizard, selecting the defaults. Reboot the system. After installing fix pack 1 for WebSphere Application Server 5.1 and starting WebSphere Application Server, rerun the WebSphere Voice Server firststeps command and verify your installation again. Then restore the WebSphere Voice Server Collector. Change directory to %WAS_HOME%\bin and issue the restoreWVSCollector.bat command.
Post-installation steps
1. Confirm that the IBM HTTP Server starts successfully by entering http://localhost:80 in a browser. This should display the Welcome to IBM HTTP Server main page. If you did not configure the IBM HTTP Server as a service during the installation, refer to Starting and stopping the IBM HTTP Server on page 142 for steps to manually start the server. 2. Use the firststeps tool to verify the installation. 3. To launch the WebSphere Application Server administrative console, enter: http://your host name:9090/admin, where your host name is the fully qualified name of your host system or your host IP address, in a browser. 4. Navigate to the Voice System View panel to confirm that WebSphere Voice Server is started. Refer to the firststeps section of the WebSphere Voice Server Information Center for more details.
Results
You have now successfully installed WebSphere Voice Server and the features you selected. The Installation wizard configures the product. You are not required to perform further configuration at this time.
Installing version 5.1.3

This topic describes how to install WebSphere Voice Server 5.1.3. The following are the default installation paths for each operating system: /usr/WebSphere/VoiceServer /opt/WebSphere/VoiceServer C:\Program Files\WebSphere\VoiceServer Remember: To install or uninstall using a Linux or an AIX platform, log in as a root user. To install or uninstall using a Windows platform, log in as a user with Administrator privileges. v Accessible options for installing WebSphere Voice Server on page 69
68
v v v v v
Installing on AIX Installing on Linux on page 71 Installing on Windows Server 2003 Enterprise or Standard editions on page 73 Post-installation steps on page 76 Results on page 76
Accessible options for installing WebSphere Voice Server

The installation instructions in the following sections use the Launchpad tool. If you cannot use the Launchpad tool, you can launch the installation directly by issuing the install command: CD_mount_point/dist/install.sh CD_mount_point/dist/install.sh CD_drive\install.bat
Installing on AIX
Tip: WebSphere Application Server and any required fix packs are installed silently with WebSphere Voice Server. While existing installations of WebSphere Application Server are supported, installing WebSphere Application Server before installing WebSphere Voice Server is not necessary. 1. Prepare your operating system platform for installation by verifying that you have adequate disk space and that you have the necessary installation prerequisites before starting installation. 2. Check to see if you have the WebSM Service enabled on your machine by issuing the following command: lslpp -h sysmgt.websm.framework If the command fails, then you do not have the WebSM Service installed. Make sure that the port 9090 is commented out of the /etc/services file, and proceed to the next step. If you do have the WebSM Service enabled, disable it by issuing the following command: /usr/websm/bin/wsmserver -disable Important: There is a port number conflict between the WebSphere Application Server administrative console and WebSM service on AIX. Both use port 9090. The port 9090 conflict on AIX is a known WebSphere Application Server issue. Ensure that no other applications are using this port, and make sure that the port 9090 is commented out of the /etc/services file. 3. Insert the WebSphere Voice Server for MRCP AIX CD. The LaunchPad tool automatically launches the Installation wizard, which is an InstallShield for MultiPlatforms (ISMP) application. The Installation wizard is in English. You can also start the Installation wizard from the product CD using the LaunchPad (launchpad.sh). Always use the fully qualified path to the launchpad.sh when installing the product. The following examples show how to start the launchpad.sh when installing from the product CD: a. Place the CD in the drive. b. Issue the appropriate AIX command in order to mount the CD and start the LaunchPad tool.
69
To unmount the CD, issue the following commands:

Failing to use the correct path can cause ISMP errors that stop the installation. Tip: You can start the installation using the install.sh command in the dist directory of the CD to bypass the LaunchPad. This is required for silent installation. The rest of this procedure assumes that you are using the Installation wizard, launched from either install.sh or LaunchPad. Corresponding entries in the response file exist for every prompt that is described as part of the wizard. Review the description of the response file for more information. Comments in the file describe how to customize its options. The LaunchPad Welcome panel displays. The Getting Started Guide link in the LaunchPad is to the docs\gettingstarted_aix_en.html file on the base product CD. Choose to read the Product Overview or the Getting Started Guide, or click Install the product to continue. The system displays an Install program has been launched message. Click OK. In the InstallShield Welcome panel, click Next to continue. Click the radio button beside the I accept the terms in the license agreement message if you agree to the license agreement and click Next to continue. After you accept the licensing terms, the installation wizard checks for prerequisites and for previous versions. The Setup screen prompts you to select either a Full or Custom install. v Select the Full radio button if you are installing WebSphere Application Server and WebSphere Voice Server. Click Next to continue. v Select the Custom radio button if you are installing WebSphere Voice Server on an existing instance of WebSphere Application Server. Click Next to continue. If you are installing WebSphere Application Server for the first time, you will be prompted for the following: a. Choose installation directories for WebSphere Application Server and for the IBM HTTP Server. b. Choose a node name and a host name for WebSphere Application Server. If you already have WebSphere Application Server installed, you are prompted to choose an existing instance or to install a new instance. Click the appropriate radio button and then click Next to continue. If an incorrect location for WebSphere Application Server is displayed, enter the correct location in the field at the bottom of the panel and click Next to continue. Next, the installation wizard prompts you for the WebSphere Application Server fix pack CD to install the required WebSphere Application Server fix packs.
4.
5. 6. 7.
8.
9.
10.
Note: If you already have WebSphere Application Server installed, you will not see this prompt from the wizard. 11. Choose the destination directory for WebSphere Voice Server. Click Next to continue.
70
12. Next, the language panel for WebSphere Voice Server displays. Select the WebSphere Voice Server TTS and/or ASR capabilities that you require. Click Next to continue. 13. A summary panel displays your install selections. Click Next to start the installation process. 14. You are then prompted to provide various CDs or image locations for each component. Note: The TTS installation is lengthy. Do not cancel out of it. The Summary panel indicates when the installation is complete, and displays the installation status. 15. A summary panel of the installation with instructions to reboot appears. Click Finish to close the installation wizard. The Welcome panel appears. Choose to read the Product Overview or the Getting Started Guide, or Exit. 16. If you receive any warnings during installation, examine the /tmp/CWVOptInstall.log file to verify that there were no file system problems, or other unusual errors. If there are problems, correct them, uninstall the product, and then reinstall. If there are no problems recorded in the /tmp/CWVOptInstall.log file, verify or troubleshoot the installation in step three described in Installing WebSphere Voice Server on page 57. 17. Restart the system and run the firststeps command on page 85. The command script is located at $WVS_ROOT/firststeps/firststeps.sh, where $WVS_ROOT is a variable defined during the installation of WebSphere Voice Server. Issue the following commands at a command prompt:
The installation of WebSphere Voice Server includes configuration tuning for WebSphere Application Server and your operating system. To see tips on additional ways to improve the performance of your WebSphere Voice Server, refer to the topic on Chapter 7, Tuning, on page 145. 18. To use a Web browser with AIX, download and install the Mozilla browser from Web browsers for AIX.
Installing on Linux
Tip: WebSphere Application Server and any required fix packs are installed silently with WebSphere Voice Server. While existing installations of WebSphere Application Server are supported, installing WebSphere Application Server before installing WebSphere Voice Server is not necessary. 1. Prepare your operating system platform for installation by verifying that you have adequate disk space and that you have the necessary installation prerequisites before starting installation. 2. Insert the WebSphere Voice Server for MRCP Linux CD. The LaunchPad tool automatically launches the Installation wizard, which is an InstallShield for MultiPlatforms (ISMP) application. The Installation wizard is in English. You can also start the Installation wizard from the product CD using the LaunchPad (launchpad.sh). Always use the fully qualified path to the launchpad.sh when installing the product. The following examples show how to start the launchpad.sh when installing from the product CD on a Linux system: a. Place the CD in the drive.
71
b. Issue the appropriate command for your Linux version in order to mount the CD and start the LaunchPad tool. For Red Hat:
For SUSE:
cd / mount /media/cdrom /media/cdrom/launchpad.sh
To unmount the CD, issue the following commands: For Red Hat:
For SUSE:
cd / umount /media/cdrom
Failing to use the correct path can cause ISMP errors that stop the installation. Tip: You can start the installation using the install.sh command in the dist directory of the CD to bypass the LaunchPad. This is required for silent installation. The rest of this procedure assumes that you are using the Installation wizard, launched from either install.sh or LaunchPad. Corresponding entries in the response file exist for every prompt that is described as part of the wizard. Review the description of the response file for more information. Comments in the file describe how to customize its options. The LaunchPad Welcome panel displays. The Getting Started Guide link in the LaunchPad is to the docs\gettingstarted_linux_en.html file on the base product CD. Choose to read the Product Overview or the Getting Started Guide, or click Install the product to continue. The system displays an Install program has been launched message. Click OK. In the InstallShield Welcome panel, click Next to continue. Click the radio button beside the I accept the terms in the license agreement message if you agree to the license agreement and click Next to continue. After you accept the licensing terms, the installation wizard checks for prerequisites and for previous versions. The Setup screen prompts you to select either a Full or Custom install. v Select the Full radio button if you are installing WebSphere Application Server and WebSphere Voice Server. Click Next to continue. v Select the Custom radio button if you are installing WebSphere Voice Server on an existing instance of WebSphere Application Server. Click Next to continue. If you are installing WebSphere Application Server for the first time, you will be prompted for the following: a. Choose installation directories for WebSphere Application Server and for the IBM HTTP Server. b. Choose a node name and a host name for WebSphere Application Server. If you already have WebSphere Application Server installed, you are prompted to choose an existing instance or to install a new instance. Click the appropriate radio button and then click Next to continue.
3.
4. 5. 6.
7.
8.
72
If an incorrect location for WebSphere Application Server is displayed, enter the correct location in the field at the bottom of the panel and click Next to continue. 9. Next, the installation wizard prompts you for the WebSphere Application Server fix pack CD to install the required WebSphere Application Server fix packs. If you already have WebSphere Application Server installed, you will not see this prompt from the wizard. Note: As you are prompted to insert additional CDs throughout the installation, make sure the default location for the CD is correct in the installation wizard when prompted. For example, /mnt/cdrom. Choose the destination directory for WebSphere Voice Server. Click Next to continue. Next, the language panel for WebSphere Voice Server displays. Select the WebSphere Voice Server TTS and/or ASR capabilities that you require. Click Next to continue. A summary panel displays your install selections. Click Next to start the installation process. You are then prompted to provide various CDs or image locations for each component.
10. 11.
12. 13.
Note: The TTS installation is lengthy. Do not cancel out of it. The Summary panel indicates when the installation is complete, and displays the installation status. 14. A summary panel of the installation with instructions to reboot appears. Click Finish to close the installation wizard. The Welcome panel appears. Choose to read the Product Overview or the Getting Started Guide, or Exit. 15. If you receive any warnings during installation, examine the /tmp/CWVOptInstall.log file to verify that there were no file system problems, or other unusual errors. If there are problems, correct them, uninstall the product, and then reinstall. If there are no problems recorded in the /tmp/CWVOptInstall.log file, verify or troubleshoot the installation in step three described in Installing WebSphere Voice Server on page 57. 16. Restart the system and run the firststeps command on page 85. The command script is located at $WVS_ROOT/firststeps/firststeps.sh, where $WVS_ROOT is a variable defined during the installation of WebSphere Voice Server. Issue the following commands at a command prompt:
The installation of WebSphere Voice Server includes configuration tuning for WebSphere Application Server and your operating system. To see tips on additional ways to improve the performance of your WebSphere Voice Server, refer to the topic on Chapter 7, Tuning, on page 145. Remember: The Linux firewall must be disabled for the Interactive Voice Response (IVR) to communicate with WebSphere Voice Server.
Installing on Windows Server 2003 Enterprise or Standard editions

Tip: WebSphere Application Server and any required fix packs are installed silently with WebSphere Voice Server. While existing installations of
73
WebSphere Application Server are supported, installing WebSphere Application Server before installing WebSphere Voice Server is not necessary. Note: If you are using Service Pack 1, download and apply the WebSphere Voice Server interim fix, IC47845, from the product Support site. 1. Prepare your operating system platform for installation by verifying that you have adequate disk space and that you have the necessary installation prerequisites before starting installation. 2. Insert the WebSphere Voice Server for MRCP Windows Server 2003 CD. The LaunchPad tool automatically launches the installation wizard. If necessary, you can also start the installation wizard from the product CD by double-clicking the launchpad.exe file. This action launches the installation wizard in English. Tip: You can start the installation using the install.bat command in the base directory of the CD to bypass the LaunchPad. This is required for silent installation. The rest of this procedure assumes that you are using the Installation wizard, launched from either install.bat or LaunchPad. Corresponding entries in the response file exist for every prompt that is described as part of the wizard. Review the description of the response file for more information. Comments in the file describe how to customize its options. 3. The LaunchPad Welcome panel displays. The Getting Started Guide link in the LaunchPad is to the docs\gettingstarted_win_en.html file on the base product CD. Choose to read the Product Overview or the Getting Started Guide, or click Install the product to continue. 4. The system displays an Install program has been launched message. Click OK. 5. In the InstallShield Welcome panel, click Next to continue. 6. Click the radio button beside the I accept the terms in the license agreement message if you agree to the license agreement and click Next to continue. After you accept the licensing terms, the installation wizard checks for prerequisites and for previous versions. 7. The Setup screen prompts you to select either a Full or Custom install. v Select the Full radio button if you are installing WebSphere Application Server and WebSphere Voice Server. Click Next to continue. v Select the Custom radio button if you are installing WebSphere Voice Server on an existing instance of WebSphere Application Server. Click Next to continue. 8. If you are installing WebSphere Application Server for the first time, you will be prompted for the following: a. Choose installation directories for WebSphere Application Server and for the IBM HTTP Server. b. Configure both WebSphere Application Server and IBM HTTP Server as a service. c. Enter a logon user ID and a password. Make sure that the logon user ID belongs to the administrator group and has the following advanced user rights: v Act as part of the operating system v Log on as a service
74
Important: When installing WebSphere Application Server as a Windows service, do not use a user ID that contains spaces. A user ID with spaces cannot be validated, and the user is not allowed to proceed with the installation. To work around this problem, install with a user ID that does not contain spaces, or do not choose to install services. d. Choose a node name and a host name for WebSphere Application Server. If you already have WebSphere Application Server installed, you are prompted to choose an existing instance or to install a new instance. Click the appropriate radio button and then click Next to continue. If an incorrect location for WebSphere Application Server is displayed, enter the correct location in the field at the bottom of the panel and click Next to continue. 9. Next, the installation wizard prompts you for the WebSphere Application Server fix pack CD to install the required WebSphere Application Server fix packs. Note: If you already have WebSphere Application Server installed, you will not see this prompt from the wizard. Choose the destination directory for WebSphere Voice Server. Click Next to continue. Next, the language panel for WebSphere Voice Server displays. Select the WebSphere Voice Server TTS and/or ASR capabilities that you require. Click Next to continue. A summary panel displays your install selections. Click Next to start the installation process. You are then prompted to provide various CDs or image locations for each component.
10. 11.
12. 13.
Note: The TTS installation is lengthy. Do not cancel out of it. The Summary panel indicates when the installation is complete, and displays the installation status. 14. A summary panel of the installation with instructions to reboot appears. Click Finish to close the installation wizard. The Welcome panel appears. Choose to read the Product Overview or the Getting Started Guide, or Exit. 15. If you receive any warnings during installation, examine the %TEMP%\CWVOptInstall.log file to verify that there were no file system problems, or other unusual errors. If there are problems, correct them, uninstall the product, and then reinstall. If there are no problems recorded in the %TEMP%\CWVOptInstall.log file, verify or troubleshoot the installation in step three described in Installing WebSphere Voice Server on page 57. 16. Restart the system and run the firststeps command on page 85. The command script is located at %WVS_ROOT%\firststeps\firststeps.bat, where %WVS_ROOT% is a variable defined during the installation of WebSphere Voice Server. Issue the following commands at a command prompt:
cd %WVS_ROOT%\firststeps firststeps.bat
The installation of WebSphere Voice Server includes configuration tuning for WebSphere Application Server and your operating system. To see tips on additional ways to improve the performance of your WebSphere Voice Server, refer to the topic on Chapter 7, Tuning, on page 145.
75
Post-installation steps
1. Confirm that the IBM HTTP Server starts successfully by entering http://localhost:80 in a browser. This should display the Welcome to IBM HTTP Server main page. If you did not configure the IBM HTTP Server as a service during the installation, refer to Starting and stopping the IBM HTTP Server on page 142 for steps to manually start the server. 2. Use the firststeps tool to verify the installation. 3. To launch the WebSphere Application Server administrative console, enter: http://your host name:9090/admin, where your host name is the fully qualified name of your host system or your host IP address, in a browser. 4. Navigate to the Voice System View panel to confirm that WebSphere Voice Server is started. Refer to the firststeps section of the WebSphere Voice Server Information Center for more details.
Results
You have now successfully installed WebSphere Voice Server and the features you selected. The Installation wizard configures the product. You are not required to perform further configuration at this time.
Installing silently
This topic describes how to perform a silent installation of the product as the root user on an AIX or Linux system, or as an administrative user on a Windows 2003 operating system.

Use this procedure to perform a silent installation of WebSphere Voice Server. A silent installation uses the installation program to install the product. The silent installation will read all of your installation options from a file that you customize. Silent installation cannot run from the CD media. You will need to copy the content of each CD to a corresponding directory (one directory for each CD). For example, you can create directories named Disk1, Disk2, and Disk3. Make sure you customize the response file to indicate the same directories you copied onto your hard disk drive. Important: Note the restrictions on the silent install in the paragraph below. A silent installation uses the -options responsefile parameter, which causes the Installation wizard to read your responses from the options response file, instead of from the interactive user interface. You must customize the response file before installing silently. After customizing the file, you must issue the command to silently install. Silent installation is particularly useful if you install the product often or if you are installing from a remote command prompt.
76
Steps for this task

Remember: To install or uninstall using a Linux or an AIX platform, log in as a root user. To install or uninstall using a Windows platform, log in as a user with Administrator privileges. 1. Copy the response file to your disk drive and customize it. The name of the original file is wvs51response.txt, and it is located in the cdrom/responsefilesample directory. 2. On your local hard disk drive image, navigate to the install script, and use the following command for silent installation: For AIX and Linux:
./install.sh -silent -options full_path_to_response_file responsefile.txt
For Windows:
install.bat -silent -options full_path_to_response_file responsefile.txt
Note: The responsefile.txt is your customized copy of the wvs51response.txt, located in the cdrom/responsefilesample directory. The full path is required to your customized copy of the response file. The log.txt file is in the logs directory in the root of the installation path of WebSphere Voice Server. For reference, the default installation paths are the following: /usr/WebSphere/VoiceServer /opt/WebSphere/VoiceServer C:\Program Files\WebSphere\VoiceServer
Results
You can view the log file to verify the install is complete. Examine the log.txt file for lines similar to these:
(Jun 30, 2004 2:53:17 PM), CWVIN, com.ibm.wps.install .ExternalCommandAction$OutputWatcher, msg2, StdOut: ADMA6016I: Add to workspace adminconsole.war/WEB-INF/struts-config_1_0.dtd (Jun 30, 2004 2:53:17 PM), CWVIN, com.ibm.wps.install.ExternalCommandAction$OutputWatcher, msg2, StdOut: ADMA6017I: Saved document /opt/WebSphere/AppServer/wstemp/fd76e74534/workspace/cells/ngbuild1/ applications/adminconsole.ear/deployments/adminconsole/adminconsole.war/ META-INF/MANIFEST.MF (Jun 30, 2004 2:53:17 PM), CWVIN, com.ibm.wps.install.ExternalCommandAction$OutputWatcher, msg2, StdOut: ADMA6016I: Add to workspace adminconsole.war/META-INF/MANIFEST.MF (Jun 30, 2004 2:53:17 PM), CWVIN, com.ibm.wps.install.ExternalCommandAction$OutputWatcher, msg2, StdOut: ADMA6017I: Saved document /opt/WebSphere/AppServer/wstemp/fd76e74534/workspace/cells/ ngbuild1/applications/ adminconsole.ear/deployments/adminconsole/adminconsole.war/ WEB-INF/web.xml (Jun 30, 2004 2:53:17 PM), CWVIN, com.ibm.wps.install.ExternalCommandAction$OutputWatcher, msg2, StdOut: ADMA6016I: Add to workspace adminconsole.war/WEB-INF/web.xml (Jun 30, 2004 2:53:17 PM),
77
CWVIN, com.ibm.wps.install.ExternalCommandAction$OutputWatcher, msg2, StdOut: ADMA5037I: Starting backup of app at /opt/WebSphere/AppServer/wstemp/fd76e74534/workspace/cells/ngbuild1/ applications/adminconsole.ear (Jun 30, 2004 2:53:29 PM), CWVIN, com.ibm.wps.install.ExternalCommandAction$OutputWatcher, msg2, StdOut: ADMA5038I: Completed backup of app at /opt/WebSphere/AppServer/wstemp/fd76e74534/workspace/cells/ngbuild1/ applications/adminconsole.ear/adminconsole.ear (Jun 30, 2004 2:53:29 PM), CWVIN, com.ibm.wps.install.ExternalCommandAction$OutputWatcher, msg2, StdOut: ADMA6011I: Deleting directory tree /tmp/app_fd76e74937 (Jun 30, 2004 2:53:44 PM), CWVIN, com.ibm.wps.install.ExternalCommandAction, msg2, Return code = 0 (Jun 30, 2004 2:53:44 PM), CWVIN, com.ibm.wps.install.ExternalCommandAction, msg2, Executing command: completed (Jun 30, 2004 2:53:45 PM), CWVIN, com.ibm.wps.install.DeleteDirectoryOnExitAction, msg1, Deleting directory tree on exit: /tmp/wvslaunch
This indicates that you have successfully installed the product. Related information Customizing the Network Deployment options response file You can customize the response file to silently install the WebSphere Application Server Network Deployment product from a command line without the Installation wizard.
Installing additional languages

This topic describes the steps to take if you want to install additional automatic speech recognition (ASR) or text-to-speech (TTS) languages on an existing installation. 1. Start the Launchpad.exe file or the launchpad.sh file to begin the installation program. 2. The LaunchPad Welcome panel displays. Click Install the product to continue. 3. The system displays an Install program has been launched message. Click OK. 4. In the InstallShield Welcome panel, click Next to continue. 5. Click the radio button beside the I accept the terms in the license agreement message if you agree to the license agreement and click Next to continue. After you accept the licensing terms, the installation wizard checks for prerequisites. 6. The language panel for WebSphere Voice Server displays. Select the WebSphere Voice Server TTS and/or ASR capabilities that you require. Click Next to continue. Tip: You do not need to reselect any languages that are currently marked as installed. 7. A summary panel displays your install selections. Click Next to start the installation process. 8. You are then prompted to provide various CDs or image locations for each component. Note: The TTS installation is lengthy. Do not cancel out of it.
78
The Summary panel indicates when the installation is complete, and displays the installation status. 9. A summary panel of the installation with instructions to reboot appears. Click Finish to close the installation wizard. 10. After restarting your machine, run the firststeps command on page 85 to make sure that the product and any newly installed ASR or TTS languages are installed correctly.
Installing interim fixes and fix packs for the product

This topic describes the proper procedure for installing any necessary fix packs for WebSphere Voice Server or its components.
WebSphere Application Server 5.1.1 fix pack required for WebSphere Voice Server 5.1.1 and 5.1.2
Remember: If you are using WebSphere Voice Server version 5.1.3, this fix pack was applied during your WebSphere Voice Server installation. Obtain fix pack 1 for WebSphere Application Server Version 5.1 Base and Network Deployment from the WebSphere Application Server support site and install it on your Linux and Windows systems using the steps below. 1. Locate WebSphere Application Server 5.1 fix pack 1 and download the package appropriate for your Linux or Windows operating system, either Linux on Intel Base or Windows Base. Note: If you are applying fix pack 1 to your Deployment Manager installation, download the Linux on Intel ND or Windows ND package. 2. Create a temporary directory for the fix pack that does not contain any spaces in the name. For example:
cd /tmp mkdir WAS511
3. Copy the fix pack file that you downloaded, was51_fp1_linux.tar.gz or was51_fp1_win.zip, to the new directory and extract its contents. For Linux, unzip and untar the file. For example:
gunzip was51_fp1_linux.tar.gz tar -xvf was51_fp1_linux.tar
4. Ensure that WebSphere Application Server and IBM HTTP Server are stopped. 5. Set up the command line. For Linux:
cd WAS_installation_directory/bin . ./setupCmdLine.sh
For Windows:
cd WAS_installation_directory\bin setupCmdLine.bat
WAS_installation_directory is the installation directory where you installed WebSphere Application Server. When you run the setupCmdLine command, the $WAS_HOME variable is defined for this location. 6. Change to the fix pack directory and start the update wizard to install the fix pack. For Linux:
79
cd /tmp/WAS511/ ./updateWizard.sh
For Windows:
cd c:\temp\WAS511/ updateWizard.bat
7. Install the fix pack by following the instructions in the wizard and selecting the defaults. 8. Reboot the system. 9. After installing WebSphere Application Server 5.1 fix pack 1 and startingWebSphere Application Server, rerun the firststeps command on page 85 and verify your installation again. 10. Then restore the WebSphere Voice Server Collector. a. Change the directory to $WAS_HOME/bin if you are using a Linux platform or %WAS_HOME%\bin if you are using a Windows platform. b. Issue the ./restoreWVSCollector.sh command if you are using a Linux platform or the restoreWVSCollector.bat if you are using a Windows platform.
Installing deployment manager

If you decide to deploy a multiple machine environment, you need to install the deployment manager (WebSphere Application Server Network Deployment) on an administrative system and then install WebSphere Voice Server administrative console extensions on this server. To complete this task, use the following steps: 1. Begin the installation of WebSphere Application Server Network Deployment by inserting the Deployment Manager CD into the CD-ROM drive. 2. Start the LaunchPad program (launchpad.sh for AIX or Linux, Launchpad.exe for Windows) and then select Install Deployment Manager. 3. Follow the instructions in the installation wizard. For detailed step-by-step instructions, refer to the following topics in the WebSphere Application Server Network Deployment Information Center: Installing Network Deployment on AIX platforms Installing Network Deployment on Linux platforms Installing Network Deployment on Windows platforms 4. Next, update WebSphere Application Server Network Deployment to the 5.1.1.6: WebSphere Application Server Version 5.1.1 Cumulative Fix 6 (or later). You can get the update from the support site, or from the CD, included with WebSphere Voice Server version 5.1.3. The CD names are as follows: WebSphere Application Server 5.1 Deployment Manager for WebSphere Voice Server 5.1.3 Fix CD AIX WebSphere Application Server 5.1 Deployment Manager for WebSphere Voice Server 5.1.3 Fix CD Linux WebSphere Application Server 5.1 Deployment Manager for WebSphere Voice Server 5.1.3 Fix CD Windows Note: If you use the download site, be sure to download the appropriate ND and SDK packages appropriate for your operating system: AIX - ND, AIX - SDK
80
Linux - ND, Linux - SDK Windows - ND, Windows - SDK 5. Install the WebSphere Voice Server administrative console extensions.
Installing WebSphere Voice Server console extensions on a deployment manager

Use the following steps to install WebSphere Application Server Network Deployment and the WebSphere Voice Server administrative console extensions. To create a deployment manager, install WebSphere Application Server Network Deployment. Note: When installing WebSphere Voice Server console extensions on a deployment manager with physical CD media, a LaunchPad program starts automatically if autorun is enabled on your system. You cannot install console extensions on a deployment manager using the LaunchPad. Use the following instructions. To add the WebSphere Application Server administrative console extensions to the deployment manager: 1. Navigate to the console installation program located in the admin directory of the WebSphere Voice Server base product CD. For AIX and Linux the installation program is install.sh. For Windows the installation program is install.bat. 2. Launch the installation manually using a fully-qualified command. For example: /mount/cdrom/admin/install.sh /mount/cdrom/admin/install.sh cdrom:\admin\install.bat Note: Mount the CD-ROM drive on Linux and AIX platforms. Consult your platform installation procedure for instructions on mounting CD-ROM drives. The installation is in English only. On the welcome panel, click Next to start the installation. If you agree, to the license agreement panel, select I accept the terms in the license agreement then click Next. Enter a target directory for the installation and click Next. The installation wizard will check for previous versions. On the summary panel, review the information and click Next to install the product. When the installation is complete, the wizard displays the status of the installation. Click Finish to close the installation wizard and restart the server.
3. 4. 5. 6. 7.
See Uninstalling Network Deployment for uninstallation instructions for WebSphere Application Server Network Deployment and the WebSphere Voice Server administrative console extensions. Related information Installing deployment manager on page 80
81
Installing the WebSphere Edge Component: Load Balancer

Remember: The Genesys Voice Platform (GVP) 7.0.3 supports the WebSphere Edge Component Load Balancer for automatic speech recognition (ASR) and text-to-speech (TTS). The GVP NE 6.5.4 and EE 6.5.5 cannot be used with the Load Balancer for ASR. However, when using GVP NE 6.5.4 and EE 6.5.5 with the IBM TTS Connector, the Load Balancer is fully supported for TTS. See Configuring the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server for detailed instructions. Use the following steps to download and install Load Balancer on your supported operating system: 1. Confirm that Load Balancer 5.1, available within the WebSphere Voice Server product set, has been installed previously. The installation of this version of Load Balancer is required to retrieve the license file needed for future service releases of Load Balancer. 2. Refer to the installation procedure for Load Balancer service releases for information about how to copy the required license file. 3. If you have WebSphere Voice Server version 5.1.1 or 5.1.2, download the Edge Server components for WebSphere Application Server at the support site. If you have WebSphere Voice Server version 5.1.3, the Load Balancer fixes are included with the product CDs. For reference, the names of the CDs are as follows: WebSphere Application Server Edge Component AIX Fixes for use with WebSphere Voice Server version 5.1.3 WebSphere Application Server Edge Component Linux Fixes for use with WebSphere Voice Server V5.1.3 WebSphere Application Server Edge Component Windows Fixes for use with WebSphere Voice Server V5.1.3 Select the appropriate Load Balancer file to download. eLBAIX51-5.1.1.51.tar.Z eLBLX-5.1.1.22-i386.tar eLBW2K-5.1.1.41.exe 5. Use the instructions in the WebSphere Edge Server technote, Installation procedures for Load Balancer service releases. 6. Refer to the Load Balancer Administration Guide for detailed steps on installing and configuring the Load Balancer. 7. Configure the server machines for your operating system: Follow the instructions in the Edge Components information center by going to Load Balancer Administration Guide Dispatcher component Configuring the Dispatcher component Setting up server machines for load balancing to set the loopback device. Update the iptables. Follow the instructions in the Edge Components information center by going to Load Balancer Administration Guide Dispatcher component Configuring the Dispatcher component Setting up server machines for load balancing to set the loopback device.
4.
82
Customizing the options response file

You can customize the response file to silently install the WebSphere Application Server Network Deployment product from a command line without the Installation wizard. For details on how to customize the response file, see Customizing the Network Deployment options response file.
Installation: Resources for learning

Use the WebSphere Application Server 5.1.x Information Center to find relevant supplemental information about installation.
Installation tips
This topic describes tips specific to installing WebSphere Voice Server. Refer to the following links for detailed tips for installation: v Non-ASCII special characters are not supported in the name of the installation directory v Verifying the system cp command when using emacs or other freeware
Using the installation verification test

Before you begin
If you are using WebSphere Voice Server versions 5.1, 5.1.1, or 5.1.2, do the following before running the installation verification test (IVT): 1. Install the product. 2. Modify the startup profile. 3. Reboot your machine. If you are using WebSphere Voice Server version 5.1.3, do the following before running the IVT program: 1. Install the product. 2. Reboot your machine.

The IVT program scans product log files for errors and verifies the core functionality of the product installation. Remember: To install or uninstall using a Linux or an AIX platform, log in as a root user. To install or uninstall using a Windows platform, log in as a user with Administrator privileges. v Steps for this task on all platforms v Sample output on page 84
Steps for this task on all platforms

1. Reboot the system after the installation process completes successfully. Note: If you try to verify the installation without rebooting, the system will not pass.
83
2. Open a terminal window and change directory to where the firststeps script is located: $WVS_ROOT/firststeps/firststeps.sh $WVS_ROOT/firststeps/firststeps.sh %WVS_ROOT%\firststeps\firststeps.bat 3. If you are using WebSphere Voice Server version 5.1, 5.1.1, or 5.1.2, you must select Start the Server before selecting Verify Installation. Selecting to verify the installation for WebSphere Voice Server does not automatically start the server on versions 5.1, 5.1.1, and 5.1.2. Important: Do not select any other options from the First Steps panel after selecting the Start the server option until the server is started. After selecting the Start the server option, an output screen displays with status messages. The success message informs you that the server is open for business (Server your_server_name open for e-business). Then the menu item changes to Stop the server (on WebSphere Voice Server versions 5.1, 5.1.1, and 5.1.2). The server must be running before proceeding to the next step. If you are using WebSphere Voice Server version 5.1.3, skip to the next step. The server is automatically started on WebSphere Voice Server version 5.1.3. 4. Confirm the server is running and select Verify Installation. The IVT runs a series of verification tests and reports the results in the First Steps output window. Do not select any other First Steps panel options while the installation verification is running. The WebSphere Voice Server IVT program validates that your ASR language and TTS voices are installed correctly.
Sample output
Below is a sample of a successful IVT report. Note that the language and voice name would be different based on your installation choices.
WebSphere Voice Server Installation Verification Test - Begin ================================================================ Starting WebSphere Application Server if its not already running ====================== Verification for Text-to-Speech voice: Andrew ============== [05/09/29 15:06:39:844 EDT] Thread-1 sysmgmt A CWVSY1002I: WVS installation verification has started. [05/09/29 15:06:39:844 EDT] Thread-1 sysmgmt A CWVSY1003I: WVS installation verification executing for TTS voice Andrew ... [05/09/29 15:06:41:625 EDT] Thread-1 sysmgmt A CWVSY1005I: WVS installation verification is validating results ... [05/09/29 15:06:41:625 EDT] Thread-1 sysmgmt A CWVSY1004I: WVS installation verification worked successfully. [05/09/29 15:06:41:625 EDT] Thread-1 sysmgmt A CWVSY1008I: WVS installation verification successfully created a sample audio file, named Andrew.au and located in the $WVS_ROOT/ivt directory. [05/09/29 15:06:41:625 EDT] Thread-1 sysmgmt A CWVSY1006I: WVS installation verification is complete. ====================== Verification for Text-to-Speech voice: Allison ============== [05/09/29 15:06:42:172 EDT] Thread-1 sysmgmt A CWVSY1002I: WVS installation verification has started. [05/09/29 15:06:42:172 EDT] Thread-1 sysmgmt A CWVSY1003I: WVS installation verification executing for TTS voice Allison ... [05/09/29 15:06:43:828 EDT] Thread-1 sysmgmt A CWVSY1005I: WVS installation verification is validating results ... [05/09/29 15:06:43:828 EDT] Thread-1 sysmgmt A CWVSY1004I: WVS installation verification worked successfully. [05/09/29 15:06:43:828 EDT] Thread-1 sysmgmt A CWVSY1008I: WVS installation verification successfully created a sample audio file, named Allison.au and located in the $WVS_ROOT/ivt directory. [05/09/29 15:06:43:828 EDT] Thread-1 sysmgmt A CWVSY1006I: WVS installation verification is complete. ====================== Verification for Automatic Speech Recognition Language: en_US ===
84
[05/09/29 15:06:44:391 EDT] Thread-1 sysmgmt A CWVSY1002I: WVS installation verification has started. [05/09/29 15:06:44:391 EDT] Thread-1 sysmgmt A CWVSY1003I: WVS installation verification executing for ASR language en_US ... [05/09/29 15:06:49:062 EDT] Thread-1 sysmgmt A CWVSY1005I: WVS installation verification is validating results ... [05/09/29 15:06:49:062 EDT] Thread-1 sysmgmt A CWVSY1004I: WVS installation verification worked successfully. [05/09/29 15:06:49:062 EDT] Thread-1 sysmgmt A CWVSY1006I: WVS installation verification is complete. ===================================================================================== WebSphere Voice Server Installation Verification Test - End Please check the log shown above for any unsuccessful test results. If any are reported, see the Troubleshooting section of the Infocenter for further information.
What to do next
Return to Installing WebSphere Voice Server to continue.
Installation verification test command

The ivt command starts the installation verification test (IVT) program. The IVT program starts the deployment manager process automatically if the process is not already running. Once the process initializes, the IVT runs a series of verification tests and reports pass or fail status in a console window. You can start the IVT program from the command line or from the First Steps menu. Because the WebSphere Voice Server IVT does not automatically start the server on versions 5.1, 5.1.1, and 5.1.2, you must start the server before running the IVT program. If you are using on WebSphere Voice Server versions 5.1.3, the server is automatically started. Note: The installation verification test is not supported on node names that contain double-byte characters. The location of the IVT script for each operating system is as follows: $WVS_ROOT/firststeps/ivt.sh $WVS_ROOT/firststeps/ivt.sh %WVS_ROOT%\firststeps\ivt.bat Log results of the ivt command are in the /logs/ivt.log file, located in the root of the WebSphere Voice Server installation path.
firststeps command
The firststeps command starts the First Steps tool. First Steps is a post-installation ease-of-use tool for directing WebSphere Voice Server elements from one place. Options dynamically appear on the First Steps panel, depending on features you install. With all options present, you can use First Steps to start and stop the server (on WebSphere Voice Server versions 5.1, 5.1.1, and 5.1.2), verify the installation, or access the information center. First Steps must be started by the command line once the system has been rebooted, after the end of installation. If you try to verify the installation without rebooting, the system will not pass. The location of the First Steps script for each operating system is listed in the Summary panel when you finish your installation. The paths are also listed here:
85
$WVS_ROOT/firststeps/firststeps.sh $WVS_ROOT/firststeps/firststeps.sh %WVS_ROOT%\firststeps\firststeps.bat
Steps for this task on all platforms

1. If you are using WebSphere Voice Server version 5.1, 5.1.1, or 5.1.2, you must select Start the Server before selecting Verify Installation. Selecting to verify the installation for WebSphere Voice Server does not automatically start the server on versions 5.1, 5.1.1, and 5.1.2. Important: Do not select any other options from the First Steps panel after selecting the Start the server option until the server is started. After selecting the Start the server option, an output screen displays with status messages. The success message informs you that the server is open for business (Server your_server_name open for e-business). Then the menu item changes to Stop the server. If you are successful in starting the server, then proceed to verifying the installation if this is the first time you have used the First Steps panel since installing the product. If you are using WebSphere Voice Server version 5.1.3, skip to the next step. The server is automatically started on WebSphere Voice Server version 5.1.3. 2. Click Verify Installation to make sure that your installation is completely successful. 3. If you do receive failures, refer to Troubleshooting installation problems on page 475.
Installing the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server
In order to use IBM WebSphere Voice Server Text-to-Speech (TTS) on the Genesys Voice Platform (GVP) NE 6.5.4 or EE 6.5.5, the IBM TTS Connector must be installed. The IBM TTS Connector is no longer needed with GVP 7.0.3. The steps in this topic are not required if you are using only automatic speech recognition (ASR).
Before you begin

Read through this topic and its related topics to prepare for installation and to make yourself familiar with installation options, before you start to use the installation tools. The IBM TTS Connector for GVP for WebSphere Voice Server supports the following platform versions: v GVP for Windows 2000 Version EE 6.5.5 with Release Number 6.5.500.73 v GVP for Windows 2000 Version NE 6.5.4 Read these topics before installing the product: v Installation prerequisites
Preparing for installation

1. Log on to your Windows GVP machine as Administrator.
86
Stop the Watchdog service. a. Open the Services dialog. b. Select Start Control Panel Administration Tools Services. c. Right-click the Watchdog service and select Stop. 3. Insert the IBM TTS Connector for GVP for WebSphere Voice Server CD into your CD-ROM drive. 4. Add a fully-qualified host name entry to the Windows GVP hosts file for WebSphere Voice Server. For example, navigate to the C:\WinNT\System32\ Drivers\etc\hosts file and add the following entry: 2.
34.34.130.1 wvs.ibm.com wvs
Installing
1. To begin the software installation, verify that the setupTTSConn.exe file is in the root directory of the IBM TTS Connector CD. 2. Then, either double-click the setupTTSConn.exe file, or launch the file from a command window. To use the command window, go to Start Programs Accessories Command Prompt and change directory to the letter containing the CD-ROM drive. 3. Follow the instructions on the installation panels. Note: For the Media Resource Control Protocol (MRCP) DNS value on the User Input panel, enter the fully-qualified host name of your WebSphere Voice Server. For example, setmrcp.wvs.ibm.com. Next type the number of channels the male voice will use, and the number of channels the female voice will use. Then select the TTS languages. Refer to Estimating resources for TTS engines on page 47 to ensure you allocated the proper number of channels. 4. When the TTS registration dialog appears after the Post-Install Summary panel, enter the following information: v TTS Name: IBM v Language Code: en-US For example: en-AU is Australian English fr-CA is Canadian French de-DE is German ja-JP is Japanese es-MX is Latin American Spanish zh-CN is Simplified Chinese en-GB is UK English en-US is US English v TTS Engine Language Code: en-US (this must match what you selected in step 3) v DLL Name: IBMTextToSpeech.DLL 5. Check the box for Male or Female Voice to reflect what you are using. 6. Enter the following information: v Voice Name: male (or female, matching the Genesys male/female panel) v Number of Channels: 16 (this number must match what you selected in step 3)
87
7.
8. 9. 10. 11. 12.
13.
Note: If you are installing another voice, repeat steps 5 and 6 and for your additional voice. Leave the Server Address and Server Port fields blank. If GVP NE is used and Directory Services are enabled, enter the Lightweight Directory Access Protocol (LDAP) user name and password. If you are not sure if Directory Services are enabled, check with your system administrator. If you have an LDAP server enabled, then entering the user name and password in this registration screen will register the IBM TTS Connector in the GVP configuration. Click OK. Repeat steps 4 through 8 for each language you selected in the earlier IBM TTS Connector registration panel. Click Yes when asked if you really want to register. Click Close. Restart the GVP: a. Open the Services dialog by going to Start Control Panel Administration Tools Services. b. Right-click the Watchdog service and select Restart. When the restarting progress bar closes, make a call to confirm that TTS is working properly.
If you need to update any of these fields after the installation process, refer to Configuring WebSphere Voice Server to support Genesys on page 112 for detailed instructions.
Prerequisites for the IBM TTS Connector for Genesys Voice Platform
This section lists any installation prerequisites for the IBM Text-to-Speech (TTS) Connector for Genesys Voice Platform (GVP). Before installing the TTS Connector on your Windows platform, GVP must be installed on the system.
Uninstalling the product

This task describes how to uninstall WebSphere Voice Server and its related products. Important: Before uninstalling, verify that you have no open Web browsers that are accessing the administrative console. Also if you have saved any scripts in the installation paths of WebSphere Voice Server or WebSphere Application Server, they may be deleted when you uninstall the products. Finally, if you installed WebSphere Voice Server 5.1.3.1, uninstall the fix pack before uninstalling 5.1.3. Remember: To install or uninstall using a Linux or an AIX platform, log in as a root user. To install or uninstall using a Windows platform, log in as a user with Administrator privileges. To uninstall WebSphere Voice Server, use the following steps: 1. Navigate to the root directory and start the uninstallation wizard.
88
$WVS_ROOT/uninstall/uninstall.sh $WVS_ROOT/uninstall/uninstall.sh %WVS_ROOT%\uninstall\uninstall.bat 2. Follow the screen options to uninstall WebSphere Voice Server, WebSphere Application Server, and your text-to-speech (TTS) and automatic speech recognition (ASR) languages. Note: The WebSphere Application Server may not appear in the uninstallation wizard list. It can be uninstalled later using its own uninstallation procedure. For reference, the commands to uninstall WebSphere Application Server are: For AIX and Linux:
$WAS_ROOT/_uninst/uninstall
For Windows:
%WAS_ROOT\_uninst\uninstall.exe
3. After the uninstall wizard finishes, remove the VoiceServer directory. 4. If the WebSphere Voice Server uninstallation program also uninstalled the WebSphere Application Server then remove the $WAS_HOME directory, such as /opt/WebSphere/AppServer. 5. Check for the entries listed in the vpd.properties file. Important: For users of WebSphere Voice Server 5.1.2 and earlier versions on Linux, remove the wvs_install_path/wvs.setenv line from your shells startup profile. For AIX users, the tuning changes applied to the root user will not be removed when you uninstall WebSphere Voice Server.
Uninstalling TTS and ASR languages

This topic describes the steps to take if you want to uninstall a single automatic speech recognition (ASR) or text-to-speech (TTS) language. Remember: To install or uninstall using a Linux or an AIX platform, log in as a root user. To install or uninstall using a Windows platform, log in as a user with Administrator privileges. 1. Navigate to the root directory and start the uninstallation wizard. $WVS_ROOT/uninstall/uninstall.sh $WVS_ROOT/uninstall/uninstall.sh %WVS_ROOT%\uninstall\uninstall.bat The InstallShield Welcome panel displays. Click Next to continue with the uninstallation. Choose the languages you wish to remove and click Next. A summary panel displays your uninstall selections. Click Next to start the uninstallation process. A summary panel of the removed files with instructions to restart appears. Click Finish to close the uninstallation wizard. Restart your machine.
2. 3. 4. 5. 6.
89
Uninstalling Network Deployment

WebSphere Application Server Network Deployment provides an uninstaller program that you can use to uninstall the product. For details on how to uninstall Network Deployment, see Uninstalling Network Deployment in the WebSphere Application Server Network Deployment information center.
Uninstalling WebSphere Voice Server administrative console extensions

Important: The administrative console extensions must be uninstalled prior to uninstalling theWebSphere Application Server Network Deployment. Remember: To install or uninstall using a Linux or an AIX platform, log in as a root user. To install or uninstall using a Windows platform, log in as a user with Administrator privileges. Use the following steps to uninstall only the WebSphere Voice Server administrative console extensions: 1. Change directory and start the uninstallation wizard using one of the following files: $WVS_ROOT/uninstall/uninstall.sh $WVS_ROOT/uninstall/uninstall.sh %WVS_ROOT%\uninstall\uninstall.bat 2. Follow the screen options to uninstall WebSphere Voice Server administrative console extensions. 3. Click Finish to exit the wizard. 4. After the uninstall wizard finishes, remove the VoiceServer directory, if it remains.
Uninstalling base WebSphere Application Server

For details on how to use the uninstaller program to uninstall WebSphere Application Server, see Uninstalling base WebSphere Application Server in the WebSphere Application Server Network Deployment Information Center.
Manually uninstalling WebSphere Application Server

This topic describes uninstalling WebSphere Application Server on AIX, Linux, and Windows platforms.
Before you begin

Uninstall WebSphere Voice Server. Read the topic on uninstalling manually in the WebSphere Application Server Information Center.
Manually uninstalling on AIX platforms

Refer to Manually uninstalling on AIX platforms in the WebSphere Application Server Information Center.
90
Manually uninstalling on Linux platforms

Refer to Manually uninstalling on Linux platforms in the WebSphere Application Server Information Center.
Manually uninstalling on Windows platforms

Refer to Manually uninstalling on Windows platforms in the WebSphere Application Server Information Center. Related information Uninstalling Network Deployment on page 90
Uninstalling the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server
This task describes how to uninstall the IBM TTS Connector for GVP for WebSphere Voice Server. 1. First, unregister the IBM TTS Connector: a. Navigate to the cn\bin\ folder in the TTS Connector installation directory, such as C:\Connector, on your Windows machine. b. Double-click the ttsreg.exe file to launch the registration panel. c. Type IBM as the TTS Vendor and click Enter. d. Click Unregister. e. Click Close. f. Restart Watchdog. 2. Go to the Start Control Panel and launch Add or Remove Programs. 3. Select the IBM TTS Connector from the list of programs displayed and click Remove. 4. Follow the instructions on the uninstallation panels.
91
92
Chapter 5. Configuring
This topic covers the configuration of voice servers. After WebSphere Voice Server is installed, system administrators might want to modify several default settings. The Configuring topics in this documentation are geared towards your initial configuration and customization of the product, although you can adjust the product at any time. The product has been designed to require minimal setup, whenever possible. The primary interface for configuring WebSphere Voice Server is the administrative console. Below are the configuration tasks and topics for administrators of WebSphere Voice Server:
Configuring using the administrative console

Launch and log in to the administrative console. You can configure the automatic speech recognition (ASR) and text-to-speech (TTS) engines using the Configuration tab, which is displayed when you click on a voice server link on the Voice Servers panel. From this view, you can access the ASR configuration parameters and the TTS configuration parameters. For more information about the administrative console view, see Monitoring the voice server and resource status.
Configuring ASR settings

Use the following steps to configure automatic speech recognition (ASR) settings using the administrative console. Remember: Configuration changes will only take effect after the voice server is stopped and restarted.
Specifying the number of ASR engines

1. Select which voice server you want to configure (for example, wvssvr1_server1), and click that voice server hyperlink in the Voice Servers panel. 2. Click the Configuration tab. 3. Navigate to the Engines to start parameter in the ASR configuration table. Place your cursor in the input box and type the desired number of engines to start. The integer number must be zero signifying no engines started, or greater. The maximum value is specific to your application and your hardwares capacity. 4. Click Apply to save your configuration changes and remain on the specific voice server view. Click OK to save your current configuration settings and return to the Voice Servers panel.
93
Preinstalling grammars
Preinstalling grammars improves the performance of a speech application, especially if the grammar is very large. Use the following steps to specify the location of a grammar file for preinstallation: 1. Select which voice server you want to configure (for example, wvssvr1_server1), and click that voice server hyperlink in the Voice Servers panel. 2. Click the Configuration tab. 3. Navigate to the Preinstalled grammars parameter in the ASR configuration table. Place your cursor in the input box and type the path and file name of the desired grammars. Multiple grammars can be separated by commas. Grammar examples:
http://vxidev1/pub/commands.grxml
or
http://vxidev1/pub/commands.grxml, http://vxidev1/pub/drinks.grxml
4. Click Apply to save your configuration changes and remain on the specific voice server view. Click OK to save your current configuration settings and return to the Voice Servers panel.
Languages installed
The Languages installed parameter in the ASR configuration table is a read-only field that specifies the recognition languages installed on your system. The languages for ASR are determined at installation.
Configuring TTS settings

Use the following steps to configure text-to-speech (TTS) settings using the administrative console. Remember: Configuration changes will only take effect after the voice server is stopped and restarted.
Specifying the number of TTS engines

1. Select which voice server you want to configure (for example, wvssvr1_server1), and click that voice server hyperlink in the Voice Servers panel. 2. Click the Configuration tab. 3. Navigate to the Engines to start parameter in the TTS configuration table. Place your cursor in the input box and type the desired number of engines to start. The integer number must be zero signifying no engines started, or greater. The maximum value is specific to your application and your hardwares capacity. 4. Click Apply to save your configuration changes and remain on the specific voice server view. Click OK to save your current configuration settings and return to the Voice Servers panel.
Preloading lexicons
Like preloading grammars, preloading lexicons also improves the performance of a speech application. Use the following steps to specify the location of a lexicon to preload:
94
1. Select which voice server you want to configure (for example, wvssvr1_server1), and click that voice server hyperlink in the Voice Servers panel. 2. Click the Configuration tab. 3. Navigate to the Lexicon pre-loads parameter in the TTS configuration table. Place your cursor in the input box and type the path and file name of the desired lexicon. Multiple lexicons can be separated by commas. There is no specific extension for lexicon files; however, you must specify the URLs using either Hypertext Transfer Protocol (HTTP) or a local file protocol. Hypertext Transfer Protocol over Secure Socket Layer (HTTPS) is not supported. Lexicon examples:
http://www.example.com/lexicon.file
or
http://www.example.com/lexicon.file, http://www.example.com/pronunciation.file
Note: All preloaded lexicon files must be encoded as UTF-8. TTS does not support lexicons for Japanese and Simplified Chinese. 4. Click Apply to save your configuration changes and remain on the specific voice server view. Click OK to save your current configuration settings and return to the Voice Servers panel.
Default voice settings

The Default voice settings is a modifiable parameter in the TTS configuration table that specifies which TTS voices are available for you to use as the MRCP server default voice. The current default voice is shown. This voice plays in your speech application whenever TTS is used and a voice is not specified. You can modify the Default voice settings by selecting a voice from the drop-down menu of available for use voices. If a TTS speak request is missing any required voice attributes, the TTS output will replace the missing attributes with the attributes of the default voice. For examples of how to set the default voice using scripting, see Configuring default and available TTS voices on page 104.
Voice(s) installed
The Voice(s) installed parameter in the TTS configuration table is a read-only field that indicates the current TTS voice(s) and languages installed on your system. TTS voices installed on your system are presented in a list, showing the languages, genders, and voice names. The languages for TTS are determined at installation. The voices that are available for use are denoted by an asterisk (*) on the administrative console. You can use the WebSphere Voice Server administration console to view voices installed on a specific voice server, or manually query the same information using the WebSphere Application Server wsadmin scripting tool. For examples, see Viewing the installed voices on a specific server on page 105.
Configuring using scripting and command line tools

Scripting
95
This documentation provides information on creating custom scripts for configuration. These scripts only work on local servers and nodes. They cannot operate on a remote server or node. To administer a remote server, you can use the wsadmin tool connected to the Deployment Manager for the cell in which the target server or node is configured. For more details about scripting, refer to the WebSphere Application Server Network Deployment 5.1.x Information Center.
Command line tools

WebSphere Application Server provides command line tools for adding and removing voice servers from a voice server cell. To manage using command line tools, perform the following steps: 1. Open a system command prompt. 2. Change to the WAS_ROOT/bin directory. Where WAS_ROOT is the directory where you installed WebSphere Application Server. 3. Run the desired command. The command runs the requested function and displays the results on the screen. Refer to the command log file for additional information. When you use the -trace option for the command, the additional trace data is captured in the command log file. The directory location for the log files is under the default system log root directory, except for commands related to a specific server instance, in which case the log directory for that server is used. You can override the default location for the command log file using the -logfile option for the command. For additional information, see Managing using command line tools in the WebSphere Application Server Network Deployment 5.1.x Information Center.
Adding a voice server

This section gives instructions for adding a voice server using the command line interface. The topology of a multiple-server WebSphere Voice Server deployment is a collection of voice servers (nodes). Each node has one application server instance, which has the WebSphere Voice Server Enterprise Archive (EAR) installed and active. To learn more about the topology of WebSphere Voice Server, see the product overview. Use the addNode command to add a voice server to a multiple-server WebSphere Voice Server deployment. You must run this command from the WAS_ROOT/bin directory of a WebSphere Voice Server installation. Depending on the size and location of the new voice server you incorporate into the cell, this command can take a few minutes to complete.
96
Important: Be sure to back up your WebSphere Voice Server configuration before adding a voice server to a cell using the backupVoiceServerConfig command-line utility. To add a node to a deployment environment using the administrative console, see Managing nodes.
Syntax
The command syntax is:
addNode deploymgr host deploymgr port -includeapps [paramter]
The deploymgr host and deploymgr port arguments are required. The default port number is 8879 for the default Simple Object Access Protocol (SOAP) port of the deployment manager. SOAP is the default Java Management Extensions (JMX) connector type for the command.
Parameters
The following parameters are available for the addNode command: -nowait Tells the addNode command not to wait for successful initialization of the launched node agent process. -quiet Suppresses the progress information that the addNode command prints in normal mode. -logfile filename Specifies the location of the log file to which information gets written. -trace Generates trace information into a file for debugging purposes.
-replacelog By default, the addNode command appends to the existing trace file. This option causes the addNode command to overwrite the trace file. -noagent Tells the addNode command not to launch the node agent process for the new node. -username name Specifies the user name for authentication if security is enabled. Acts the same as the -user option. -user name Specifies the user name for authentication if security is enabled. Acts the same as the -username option. -password password Specifies the password for authentication if security is enabled. -conntype type Specifies the JMX connector type to use for connecting to the deployment manager. Valid types are SOAP or RMI, which stands for Remote Method Invocation. -startingport portNumber Supports the specification of a port number to use as the base port number for all node agent and Java Messaging Service (JMS) server ports created
97
during the addNode command. With this support you can control which ports are defined for these servers, rather than using the default port values. The starting port number is incremented to calculate the port number for every node agent port and JMS server port configured during the addNode command. -help -? Prints a usage statement. Prints a usage statement.
Usage scenario
The following examples demonstrate correct syntax for Linux and AIX:
./addNode.sh testhost 8879 -includeapps ./addNode.sh testhost 8879 -includeapps (from the remote node command line) ./addNode.sh deploymgr 8879 -includeapps -trace (produces the addNode.log file) ./addNode.sh host25 8879 -includeapps -nowait (does not wait for a node agent process)
The following examples demonstrate correct syntax for Windows:

addNode testhost 8879 -includeapps addNode testhost 8879 -includeapps (from the remote node command line) addNode deploymgr 8879 -includeapps -trace (produces the addNode.log file) addNode host25 8879 -includeapps -nowait (does not wait for a node agent process)
Where 8879 is the default port. Example output from a Linux machine:
[superuser@mynode bin] ./addNode.sh my_ibm_host 8879 -includeapps ADMU0116I: Tool information is being logged in file /opt/WebSphere/AppServer/logs/addNode.log ADMU0001I: Begin federation of node ibm_node with Deployment Manager at my_ibm_host:8879. ADMU0009I: Successfully connected to Deployment Manager Server: my_ibm_host:8879 ADMU0505I: Servers found in configuration: ADMU0506I: Server name: server1 ADMU2010I: Stopping all server processes for node ibm_node ADMU0510I: Server server1 is now STOPPED ADMU0024I: Deleting the old backup directory.
98
ADMU0015I: Backing up the original cell repository. ADMU0012I: Creating Node Agent configuration for node: ibm_node ADMU0014I: Adding node ibm_node configuration to cell: ibm_cell ADMU0016I: Synchronizing configuration between node and cell. ADMU0018I: Launching Node Agent process for node: ibm_node ADMU0020I: Reading configuration for Node Agent process:nodeagent ADMU0022I: Node Agent launched. Waiting for initialization status. ADMU0030I: Node Agent initialization completed successfully. Process id is: 2382 ADMU0523I: Creating queue manager for node ibm_node on server jmsserver ADMU0525I: Details of queue manager creation can be seen in the file: createMQ.ibm_node_jmsserver.log ADMU9990I: ADMU0300I: Congratulations! Your node ibm_node has been successfully incorporated into the ibm_cell cell. ADMU9990I: ADMU0306I: Be aware: ADMU0302I: Any cell-level documents from the standalone ibm_node configuration have not been migrated to the new cell. ADMU0307I: You might want to: ADMU0303I: Update the configuration on the ibm_cell Deployment Manager with values from the old cell-level documents. ADMU9990I: ADMU0003I: Node ibm_node has been successfully federated.
Removing a voice server

This section gives instructions for removing a voice server using the command line interface. The topology of a multiple-server WebSphere Voice Server deployment is a collection of voice servers (nodes). Each node has one application server instance, which has the WebSphere Voice Server Enterprise Archive (EAR) installed and active. To learn more about the topology of WebSphere Voice Server, see the product overview. Use the removeNode command to remove a voice server from a multiple-server WebSphere Voice Server deployment. You must run this command from the WAS_ROOT/bin directory of a WebSphere Voice Server installation.
99
The removeNode command only removes the node-specific configuration from the cell. This command does not uninstall any applications that were installed as the result of executing an addNode command. Such applications can subsequently deploy on additional servers in the Network Deployment cell. As a consequence, an addNode command with the -includeapps option run after a removeNode command does not move the applications into the cell because they already exist from the first addNode command. The resulting application servers added on the node do not contain any applications. To deal with this situation, add the node and use the deployment manager to manage the applications. Add the applications to the servers on the node after it is incorporated into the cell. Depending on the size and location of the new node you remove from the cell, this command can take a few minutes to complete. Note: To remove a node from a deployment environment using the administrative console, see Managing nodes.
Syntax
The command syntax is:
removeNode [parameter]
All arguments are optional.
Parameters
The following parameters are available for the removeNode command: -quiet Suppresses the progress information that the removeNode command prints in normal mode. -logfile fileName Specifies the location of the log file to which information is written. -replacelog Replaces the log file instead of appending to the current log. -trace Generates trace information into a file for debugging purposes.
-statusport portNumber Specifies that an administrator can set the port number for the node agent status callback. -username name Specifies the user name for authentication if security is enabled. Acts the same as the -user option. -user name Specifies the user name for authentication if security is enabled. Acts the same as the -username option. -password password Specifies the password for authentication if security is enabled. -force Cleans up the local node configuration regardless of whether you can reach the deployment manager for cell repository cleanup. After using the -force parameter, you can need to use the cleanupNode command on the deployment manager. -help Prints a usage statement.
100
-?
Prints a usage statement.
Usage scenario
The following example demonstrates correct syntax for Linux and AIX:
./removeNode.sh -quiet ./removeNode.sh -trace (produces the removeNode.log file)
The following example demonstrates correct syntax for Windows:

removeNode -quiet removeNode -trace (produces the removeNode.log file)
Creating custom scripts for configurations

Configuration scripts use the WVSAdminConfig object to access the repository where configuration information is stored. You can use the WVSAdminConfig object to list configuration parameters and their values. 1. Decide how you want to run the script. To save the script for repeated use, compose it in a file and run it with the wsadmin -f filename command. If you want to compose and run the script interactively, issue the wsadmin.sh or wsadmin.bat command without the -f flag. Note: The wsadmin.sh command is used for Linux and AIX operating systems and the wsadmin.bat command is used with Windows operating systems. For more information about executing scripts, see the WebSphere Application Server Network Deployment 5.1.x Information Center. 2. Change to the WAS_ROOT/bin directory. Where WAS_ROOT is the directory where you installed WebSphere Application Server. 3. Write a Jacl script that updates or queries the WebSphere Voice Server configuration values (examples below). It is recommended you write the Jacl script in a text file with a .jacl extension and run it using the method described in step 1. Example 1 for WebSphere Voice Server versions 5.1, 5.1.1, and 5.1.2. The following Jacl code uses an WVSAdminConfig script command statement to update the number of automatic speech recognition (ASR) engines:
# Query WVS configuration object full name set tmpCfg [$AdminControl queryNames *:*,type=WVSAdminConfig] # Query current number of ASR engines set asreng [$AdminControl invoke $tmpCfg get com.ibm.voice.server.recognizer.asr.en_US.pool] puts stdout "Current ASR engines: $asreng" # Update number of ASR engines $AdminControl invoke $tmpCfg put {com.ibm.voice.server.recognizer.asr.en_US.pool 2 true} # Query new number of ASR engines set asreng [$AdminControl invoke $tmpCfg get com.ibm.voice.server.recognizer.asr.en_US.pool] puts stdout "New ASR engines: $asreng"
Example 1.a for WebSphere Voice Server version 5.1.3.
101
The following Jacl code uses an WVSAdminConfig script command statement to update the number of automatic speech recognition (ASR) engines:
# Query WVS configuration object full name set tmpCfg [$AdminControl queryNames *:*,type=WVSAdminConfig] # Query current number of ASR engines set asreng [$AdminControl invoke $tmpCfg get com.ibm.voice.server.recognizer.asr.pool] puts stdout "Current ASR engines: $asreng" # Update number of ASR engines $AdminControl invoke $tmpCfg put {com.ibm.voice.server.recognizer.asr.pool 2 true} # Query new number of ASR engines set asreng [$AdminControl invoke $tmpCfg get com.ibm.voice.server.recognizer.asr.pool] puts stdout "New ASR engines: $asreng"
Example 2 The following Jacl code uses an WVSAdminConfig script to set TTS pre-lexicon load values:
# Query WVS configuration object full name set tmpCfg [$AdminControl queryNames *:*,type=WVSAdminConfig] # Query current TTS lexicon preload value set ttslex [$AdminControl invoke $tmpCfg get com.ibm.voice.server.synth.engine.preload-lexicons] puts stdout "Current TTS lexicon preloads: $ttslex" # Update TTS lexicon preload value $AdminControl invoke $tmpCfg put { com.ibm.voice.server.synth.engine.preload-lexicons file:///tmp/grammar/lexi1.lex true} # Query new TTS lexicon preload value set ttslex [$AdminControl invoke $tmpCfg get com.ibm.voice.server.synth.engine.preload-lexicons] puts stdout "New TTS lexicon preloads: $ttslex"
You should also review Modifying nested attributes with the wsadmin tool.
Configuration Service MBean

Use the WebSphere Voice Server Configuration Service managed bean (MBean), WVSAdminConfig, to invoke configuration commands and to create or change elements of the WebSphere Voice Server configuration, such as the text-to-speech (TTS) or automatic speech recognition (ASR) engine configurations. Use sample scripts putParam and getParam for managing most configuration changes or queries. However, if you need to perform customized or more involved configuration changes or queries of your WebSphere Voice Server configuration settings, you should use the WebSphere Voice Server Configuration Service MBean . The WVSAdminConfig service is started with the application server instance. Important: Start the scripting client shell wsadmin, after starting the application server. The following commands are available for the WVSAdminConfig MBean: getlist Returns a list of all the configuration technical names matching a filter string.
102
Note: The wild card (*) attribute does not have to be after a period. For example, com.ibm.voice.server.asr*. Usage: getList parameter wild card get Returns the value of the specified parameter name. Usage: get parameter put Updates the value for the specified parameter. Neither the parameter name nor the parameter value can be null. The last parameter must be true so that changes made are saved. Usage: put parameter parameter value true Remember: Configuration changes will only take effect after the voice server is stopped and restarted.
Sample code
The following sample shows you how to find and access the WVSAdminConfig MBean interface and services.
wsadmin> set wvscfg [$AdminControl queryNames *:*, type=WVSAdminConfig] WebSphere:cell=nextgen,name=WVSAdminConfig, mbeanIdentifier=cells/nextgen/ nodes/nextgen/servers/server1/serverl.xml#CustomService_1079378808193, type=WVSAdminConfig,node=nextgen,process=server1 wsadmin> $Help operations $wvscfg Operation java.util.Properties getList(java.lang.String) java.lang.String get(java.lang.String) void put(java.lang.String, java.lang.String) void put(java.lang.String, java.lang.String, boolean) void put(java.util.Properties, boolean) void backup(java.lang.String) void restore(java.lang.String) wsadmin> $AdminControl invoke $wvscfg get com.ibm.voice.server.rr.fetchthreads 11 wsadmin> $AdminControl invoke $wvscfg getList * {com.ibm.voice.server.synth.mrcp.Zhou-Hong.speech-language zh_CN} {com.ibm.voice.server.synth.mrcp.Shizuka.voice-age adult} {com.ibm.voice.server.synth.mrcp.Allison.voice-age adult} {com.ibm.voice.server.synth.mrcp.Shizuka.prosody-range 10} {com.ibm.voice.server.synth.mrcp.Daichi.prosody-rate 50} {com.ibm.voice.server.synth.mrcp.Shizuka.speech-language ja_JP} {com.ibm.voice.server.synth.mrcp.Shizuka.prosody-pitch 268} {com.ibm.voice.server.synth.mrcp.Andrew.prosody-rate 50} {com.ibm.voice.server.synth.mrcp.Zhou-Hong.prosody-range 10} {com.ibm.voice.server.rr.cachefilesize 9} {com.ibm.voice.server.synth.mrcp.Shizuka.prosody-rate 50} {com.ibm.voice.server.synth.mrcp.Zhou-Hong.prosody-pitch 137} {com.ibm.voice.server.rr.caching true} {com.ibm.voice.server.synth.mrcp.Daichi.voice-age adult} {com.ibm.voice.server.synth.mrcp.David.voice-age adult} {com.ibm.voice.server.synth.mrcp.Zhou-Hong.prosody-rate 50} {com.ibm.voice.server.synth.mrcp.Andrew.voice-age adult} . . . wsadmin> $AdminControl invoke $wvscfg put
103
{com.ibm.voice.server.rr.fetchthreads 20 true} wsadmin> $AdminControl invoke $wvscfg putProperties {{{com.ibm.voice.server.rr.fetchthreads 20} {com.ibm.voice.server.synth.pool-size 3} {com.ibm.voice.server.common.codec G711ulaw} } true}
Configuring default and available TTS voices

This topic describes how to use scripting to modify which of your text-to-speech (TTS) voices are available for use. All of your installed TTS voices are marked as available for use automatically when you install WebSphere Voice Server. In order to optimize the performance of your WebSphere Voice Server, administrators may want to configure fewer voices as available for use. TTS voices that are available for use are marked with an asterisk (*) on the administrative console. An Administrator can configure a default TTS voice for each voice server. The voice server uses the default voice parameter to assist in the selection of a TTS voice when a TTS speak request is missing the necessary voice attributes. The default voices gender, age, and language attributes can be used if a TTS speak request does not provide them explicitly. The default TTS voice name must be selected from the list of available TTS voices.
Examples of setting the TTS default voice

This example uses the putParam sample script on a Windows machine. The putParam sample on page 128 script is the preferred method for configuring the default TTS voice.
cd C:\Program Files\WebSphere\VoiceServer\samples putParam.bat com.ibm.voice.server.synth.default-voice Andrew
This example uses Jacl to run the WebSphere Application Server wsadmin scripting tool.
wsadmin> set wvscfg [$AdminControl queryNames *:*,type=WVSAdminConfig] WebSphere:cell=nextgen,name=WVSAdminConfig, mbeanIdentifier=cells/nextgen/nodes/nextgen/servers/server1/serverl.xml#CustomService_1079378808193, type=WVSAdminConfig,node=nextgen,process=server1 wsadmin> $AdminControl invoke $wvscfg put {com.ibm.voice.server.synth.default-voice Andrew true}
Examples of setting the TTS available voices

This example uses the putParam sample script on a Windows machine. The putParam sample on page 128 script is currently the only method for configuring available TTS voices.
cd C:\Program Files\WebSphere\VoiceServer\samples putParam.bat com.ibm.voice.server.synth.voices Andrew,Dawn
This example uses Jacl to run the WebSphere Application Server wsadmin scripting tool.
wsadmin> set wvscfg [$AdminControl queryNames *:*,type=WVSAdminConfig] WebSphere:cell=nextgen,name=WVSAdminConfig, mbeanIdentifier=cells/nextgen/nodes/nextgen/servers/server1/serverl.xml#CustomService_1079378808193, type=WVSAdminConfig,node=nextgen,process=server1 wsadmin> $AdminControl invoke $wvscfg put {com.ibm.voice.server.synth.voices Andrew,Dawn true}
104
Viewing the installed voices on a specific server

You can use the WebSphere Voice Server administration console to view voices installed on a specific voice server, or manually query the same information using the WebSphere Application Server wsadmin scripting tool. For example:
wsadmin> set wvsapp [$AdminControl queryNames *:*,type=WVSAdminApp] WebSphere:cell=nextgen,name=WVSAdminApp, mbeanIdentifier=cells/nextgen/nodes/nextgen/servers/server1/serverl.xml#CustomService_1079378808193, type=WVSAdminApp,node=nextgen,process=server1 wsadmin> $AdminControl invoke $wvsapp getTTSVoices Allison,Andrew,Dawn,Tyler
For more information about the installed and default voices for TTS, refer to Configuring TTS settings on page 94.
Configuring a MIME map on your customer Web server

You need to create Multipurpose Internet Mail Extensions (MIME) mappings between the associated extensions and the corresponding content type in order to set the various file types returned to WebSphere Voice Server in the HTTP header.
Associated extension .grxml .abnf .ulw .alw .wav .xml .au .pkg .lxml Content type (MIME) application/srgs+xml application/srgs audio/basic audio/x-alaw-basic audio/x-wav text/xml audio/basic application/srgs application/x-ibm-lexicon+xml
Configuration parameters
This section lists configuration parameters by categories of parameters related to the cache, automatic speech recognition (ASR) configuration, text-to-speech (TTS) configuration, and Real Time Streaming Protocol (RTSP) Bridge resources.
Cache-related parameters
WebSphere Voice Server has two different levels of caches for each server. It has a resource first level cache available to hold external files provided by the customer such as grammars, lexicon documents, and prerecorded audio files. It has a grammar second level cache available to hold the compiled grammars. These two levels of caches allow for the quick loading of grammars that are shared between ASR engines. The following parameters control the two levels of caching for each server. They identify the size and location of the caches. If you are developing your application, disable the first level cache until any application errors have been fixed.
105
Note: The parameters in this table can be updated using scripting.

Table 11. Cache-related parameters
Name Grammar (second level) Default Cache Size (MB) Grammar (second level) Cache Location com.ibm.voice.server.gm.cachename Default location of where the Compiled Grammars are stored on the local file $WVS_ROOT/gmcache Must be a valid US-ASCII directory on the local file system (optionally ending in the file separator character such as /) A positive integer value in megabytes Technical name and description com.ibm.voice.server.gm.cachefilesize Default grammar file cache capacity Default value 9 MB Valid values A positive integer value in megabytes
General (first level) com.ibm.voice.server.rr.cachefilesize Default Cache Size (MB) Default general file cache capacity
9 MB
General (first level) com.ibm.voice.server.rr.fetchexpires Default Expiration Time Default value for all resources if expiration is not provided by the fetched resource
86400000 ms
A positive integer value in milliseconds
General (first level) com.ibm.voice.server.rr.cachename Cache Location Location of the first level general cache
$WVS_ROOT/rrcache
Must be a valid US-ASCII directory on the local file system (optionally ending in the file separator character such as /) A positive integer value in megabytes
General (first level) com.ibm.voice.server.rr.cachememsize Memory Cache Size General first level memory cache capacity
2 MB
General (first level) com.ibm.voice.server.rr.caching Caching Enabled Enable/disable caching
True
True/False
ASR configuration parameters

The following parameters control the ASR engine. These parameters are also present in the ASR configuration table in the administrative console. For more information on the parameters from the console view, see Configuring ASR settings. Note: The parameters in this table can be updated using scripting or through the administrative console.
106
Table 12. ASR configuration parameters

Name Engines to start Note: The value for this parameter matches the number of Engines to start on the administrative console ASR configuration table. Technical name and description com.ibm.voice.server.recognizer.asr.en_AU.pool com.ibm.voice.server.recognizer.asr.fr_CA.pool com.ibm.voice.server.recognizer.asr.de_DE.pool com.ibm.voice.server.recognizer.asr.ja_JP.pool com.ibm.voice.server.recognizer.asr.es_MX.pool com.ibm.voice.server.recognizer.asr.zh_CN.pool com.ibm.voice.server.recognizer.asr.en_GB.pool com.ibm.voice.server.recognizer.asr.en_US.pool v en_AU is Australian English v fr_CA is Canadian French v de_DE is German v ja_JP is Japanese v es_MX is Latin American Spanish v zh_CN is Simplified Chinese v en_GB is UK English v en_US is US English Default value For WebSphere Voice Server version 5.1.3, the default value is 1. For WebSphere Voice Server versions 5.1, 5.1.1, and 5.1.2, the default value is 0. Valid values Zero (0) to the maximum number of engines that can be hosted by your hardware configuration.
com.ibm.voice.server.recognizer.asr.pool-size Number of ASR engines to be started. Zero (0) means that no ASR support for that language is configured. Grammar preloads Note: The value for this parameter matches the number of Grammar pre-loads on the administrative console ASR configuration table. com.ibm.voice.server.recognizer.grammar.preload Defines a list of grammars to be preloaded when the system is started. If the URI represents the source of the grammar, it will be compiled before it is loaded. Empty A list of URIs separated by commas. For more information, refer to Preinstalling grammars on page 94.
TTS configuration parameters

The following parameters control the TTS voice(s) installed on your voice server. The gender and name of TTS voices are not modifiable and are as follows: Canadian French Pierre adult male Louise adult female German Dieter adult male Erika adult female Japanese Daichi adult male Shizuka adult female Latin American Spanish Rafael adult male Sofia adult female Simplified Chinese Zhou-Hong adult female
107
U.K. English David adult male Dawn and Sally adult female voices Kate adult fifth generation female voice U.S. English Andrew adult male Allison and Tyler adult female voices Julie and Lisa adult fifth generation female voices Note: v The parameters in this table can be updated using scripting. v Depending on the capacity of your WebSphere Voice Server, the number of voices that can be loaded might be limited to one fifth generation voice and an additional voice. The configurable (unless otherwise indicated) parameters for these voices are listed below.
Table 13. TTS voice configuration parameters
Name Pitch Technical name and description com.ibm.voice.server.synth.mrcp.voicename.prosody-pitch Speakers base pitch in hertz (Hz) Default value Allison = 226Hz Andrew = 109Hz David = 105Hz Dawn = 200Hz Daichi = 101Hz Dieter = 98Hz Erika = 146Hz Julie = 192Hz Kate = 174Hz Lisa = 168Hz Louise = 180Hz Pierre = 124Hz Rafael = 128Hz Sally = 171Hz Shizuka = 268Hz Sofia = 182Hz Tyler = 226Hz Zhou-Hong = 137Hz Valid values 40 - 422Hz
108
Table 13. TTS voice configuration parameters (continued)

Name Pitch range Technical name and description com.ibm.voice.server.synth.mrcp.voicename.prosody-range Speakers pitch fluctuation Default value Allison = 170Hz Andrew = 126Hz Daichi = 95Hz David = 109Hz Dawn = 142Hz Dieter = 98Hz Erika = 246Hz Julie = 143Hz Kate = 224Hz Lisa = 149Hz Louise = 164Hz Pierre = 175Hz Rafael = 63Hz Sally = 224Hz Shizuka = 163Hz Sofia = 119Hz Tyler = 186Hz Zhou-Hong = 169Hz Loudness com.ibm.voice.server.synth.mrcp.voicename.prosody-volume Loudness Julie = 100 Kate = 100 Lisa = 100 Zhou-Hong = 94 All other languages = 92 0 - 100 Valid values [0-9]+Hz
109
Table 13. TTS voice configuration parameters (continued)

Name Rate Technical name and description com.ibm.voice.server.synth.mrcp.voicename.prosody-rate Speakers speaking rate in words per minute (wpm) Default value Allison = 171 Andrew = 175 Daichi = 164 David = 155 Dawn = 150 Dieter = 127 Erika = 122 Julie = 180 Kate = 173 Lisa = 187 Louise = 170 Pierre = 175 Rafael = 161 Sally = 173 Shizuka = 164 Sofia = 144 Tyler = 173 Zhou-Hong = 176 Valid values 70 - 1297
Additional parameters for the systems voices are the ones that map to the TTS configuration table in the administrative console. For more information on the parameters from the console view, see Configuring TTS settings. Note: The parameters in this table can be updated using scripting or through the administrative console.
Table 14. Additional TTS voice configuration parameters
Name Available voices Technical name and description com.ibm.voice.server.synth.voices A list of voices installed in the TTS engine. Default value Installed voice names Valid values A list of voices separated by commas, to make available to Voice Server client requests. For more information on how to configure available voices, refer to Configuring default and available TTS voices. Lexicon preload com.ibm.voice.server.synth.engine.preloadlexicons A list of lexicon files to be pre-loaded for all TTS engines. Empty A list of URIs separated by commas.
110
Table 14. Additional TTS voice configuration parameters (continued)

Name Engines to start Note: The value for this parameter matches the number of Engines to start on the administrative console TTS configuration table. Technical name and description com.ibm.voice.server.synth.pool-size Number of TTS engines to be started. Zero (0) means that no TTS support for that language is configured. Default value 1 Valid values Zero (0) to the maximum number of engines that can be hosted by your hardware configuration.
Default voice com.ibm.voice.server.synth.default-voice Note: The value for this parameter Default voice name used by all TTS engines. matches the number of Default voice settings on the administrative console TTS configuration table.
Australian English = David French Canadian = Pierre German = Dieter Japanese = Daichi LA Spanish = Rafael Simplified Chinese = Zhou-Hong UK English = David US English = Andrew
A voice name available for use. For more information on default voices, refer to Default voice settings on page 95. For more information on how to configure available voices, refer to Configuring default and available TTS voices on page 104.
RTSP parameters
These parameters define the port and URL to use when contacting the WebSphere Voice Server from the qualified compatible Interactive Voice Response (IVR) system. Note: The parameters in this table can be updated using scripting.
Table 15. RTSP parameters
Name RTSP Port Technical name and description com.ibm.voice.server.rtspbridge.port The port from which the RTSP server listens RTSP Recognizer URL RTSP Synthisizer URL RTSP Teardown com.ibm.voice.server.rtspbridge.recourl RTSP media resource URL for the recognizer com.ibm.voice.server.rtspbridge.synthurl RTSP media resource URL for the synthesizer com.ibm.voice.server.rtspbridge.oneteardown Allows the server to accept one teardown from the client in order to clean up all resources. If you set the value to true, the server will only expect one RTSP Teardown. However, if you set it to false (the default), then the server will expect an RTSP Teardown for both ASR and TTS. RTSP Socket Time Out com.ibm.voice.server.rtspbridge.sockettimeout The amount of time RTSP Bridge will wait for a request from the client before timing out while in an established session 300 A positive integer Note: Set the default to value in seconds 0 when using WebSphere Voice Server with the Genesys Voice Platform (GVP) NE 6.5.4 or EE 6.5.5. G711ulaw G711ulaw or G711alaw False True/False /media/ synthesizer Any valid relative URL /media/ recognizer Any valid relative URL Default value 554 Valid values Any available port
Default Audio Codec Type
com.ibm.voice.server.common.codec Type of audio codec
111
Configuring WebSphere Voice Server to support Genesys

Use these steps to configure WebSphere Voice Server for Genesys support. Remember: The Genesys Voice Platform (GVP) 7.0.3 supports the WebSphere Edge Component Load Balancer for automatic speech recognition (ASR) and text-to-speech (TTS). The GVP NE 6.5.4 and EE 6.5.5 cannot be used with the Load Balancer for ASR. However, when using GVP NE 6.5.4 and EE 6.5.5 with the IBM TTS Connector, the Load Balancer is fully supported for TTS. See Configuring the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server for detailed instructions. 1. Ensure that WebSphere Application Server is running. 2. Change directory to the WebSphere Application Server installation path, such as: /usr/WebSphere/AppServer/bin /opt/WebSphere/AppServer/bin C:\Program Files\WebSphere\AppServer\bin 3. Make sure that Global Security is disabled on WebSphere Application Server. The script to configure Genesys must be run with WebSphere Application Server Global Security disabled. 4. Run the ./configureGenesys.sh file for AIX or Linux, or the configureGenesys.bat file for Windows platforms. v If you are using GVP 7.0.3, pass the parameter true when you run the configureGenesys script. For example: configureGenesys.bat true v If you are using an earlier version of GVP, pass the parameter false when you run the configureGenesys script. For example: configureGenesys.bat false 5. Stop and then restart WebSphere Application Server.
Configuring WebSphere Voice Server to support WebSphere Voice Response

Use these steps to configure WebSphere Voice Server for WebSphere Voice Response support. If you have WebSphere Voice Server version 5.1.2 or earlier, you can obtain the WebSphere Voice Server Connector to configure WebSphere Voice Server to use WebSphere Voice Response from the Support site. The WebSphere Voice Server Connector script is included in the interim fixes for WebSphere Voice Server versions 5.1.1 and 5.1.2. If you have WebSphere Voice Server version 5.1.3, the script is included on your product CDs. Use the steps below to install the script on WebSphere Voice Server version 5.1.3. 1. Ensure that WebSphere Application Server is running. 2. Change directory to the WebSphere Application Server installation path, such as: /usr/WebSphere/AppServer/bin
112
/opt/WebSphere/AppServer/bin C:\Program Files\WebSphere\AppServer\bin 3. Make sure that Global Security is disabled on WebSphere Application Server. The script to configure WebSphere Voice Response must be run with WebSphere Application Server Global Security disabled. 4. Run the ./configureWVR.sh file for AIX or Linux, or the configureWVR.bat file for Windows platforms. 5. Stop and then restart WebSphere Application Server.
Configuring the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server
This section identifies the configurable parameters in the IBMTextToSpeech.ini file. The IBMTextToSpeech.ini file is located in the cn\config folder in the TTS Connector installation directory on your Windows machine. The following parameters are replaced in the IBMTextToSpeech.ini by the TTS Connector installation. Please verify that these values are accurate for your installation. v SERVERNAME v TOTALCHANNELS v LOGDIR v LANG v FEMALECHANNELS v MALECHANNELS If you change any of these parameters, you need to stop and restart Windows Watchdog in order for the new values to take effect: 1. Access Watchdog from Start Control Panel Administrative Tools Services. 2. Select the Watchdog application from the list of Services. 3. Click Stop the service. 4. Run the TTS registration dialogue as documented in the installing the TTS Connector section. 5. Then click Start the service.
Table 16. IBMTextToSpeech.ini file parameters
Parameter MrcpUrl Description The URL of the synthesizer resource on the Media Resource Control Protocol (MRCP) server. mrcpurl=rtsp://SERVERNAME /media/synthesizer Channels The number of channels of all the voices used. Channels=TOTALCHANNELS The TOTALCHANNELS variable must be equal to the sum of NumTTSEngines of all the voices used. The number must equal the number of text-to-speech (TTS) engines configured in the server. Valid values The SERVERNAME is the name of your MRCP server.
113
Table 16. IBMTextToSpeech.ini file parameters (continued)

Parameter AudioFrameSize Description The size of the audio frame in milliseconds (the jitter buffer). The Real-Time Transport Protocol (RTP) receiver accumulates this much audio before it sends it to the client. For example: AudioFrameSize=500 AudioBufferSize The size of the circular audio buffer in milliseconds. For example: AudioBufferSize=1500 Channel AllocationTimeout The amount of time a request should wait for a free channel in milliseconds. For example: ChannelAllocationTimeout=-1 LogFile The location of the log where all errors LOGDIR is the directory for log and and warnings for all channels are trace files. written. LogFile=LOGDIR \mrcplog.txt MrcpTraceDir If EnableTrace is set, then a file called MRCP-xxx.txt will be created for each channel with the messages exchanged through the RTSP/MRCP socket. MrcpTraceDir=LOGDIR EnableTrace This parameter controls the creation of the MRCP trace files and enables detailed log messages. For example: EnableTrace=0 LanguagePack List This parameter has a list with all the voice sections, separated by a semicolon. Each LanguagePackList contains the following parameters: VendorLanguage, NumTTSEngines, Language, Gender, and VoiceName. For example: LanguagePackList= LANG-female, LANG-male TTSVendor=IBM [tts_ibm_LANG-female] VoiceName=female Gender=F mrcp-Voice-gender=female VendorLanguage=LANG NumTTSEngines=FEMALECHANNELS Language=LANG mrcp-speech-language=LANG [tts_ibm_LANG-male] VoiceName=male Gender=M mrcp-Voice-gender=male VendorLanguage=LANG NumTTSEngines=MALECHANNELS Language=LANG mrcp-speech-language=LANG The LANG variable is the appropriate TTS language code, such as en-us for US English. The FEMALECHANNELS and MALECHANNELS variables are the maximum number of channels for the female voice and the maximum number of channels for the male voice, respectively. Note: Refer to Estimating resources for TTS engines on page 47 to ensure you allocated the proper number of channels. 0 is the default and logs errors; a value of 1 creates MRCP trace files. LOGDIR is the directory for log and trace files. This buffer should be at least two to four times the AudioFrameSize. This number is automatically rounded up to the next integer of frames. -1 is the default and waits forever; 0 returns immediately with error Valid values The AudioFrameSize value is recommended to be a multiple of the number of Real-Time Transport Protocol (RTP) frames in order to avoid unnecessary delays. Also it should be a divisor of the Genesys jitter buffer size.
114
Table 16. IBMTextToSpeech.ini file parameters (continued)

Parameter RequestsPerSession Description The number of SPEAK requests that are to be sent in a single MRCP session. Valid values 0 - sends SPEAK requests reusing the existing session indefinitely. No disconnect request is sent. This is the default. 1 - sends a single SPEAK request and then a disconnect. Note: If you use the IBM TTS Connector with the WebSphere Edge Component Load Balancer, set the RequestsPerSession parameter to 1.
115
116
Chapter 6. Administering
Administration is the set of tasks by which you modify, control, and monitor your voice server system for general use. The primary interface for system administration is the administrative console. However, a command line interface also provides these capabilities. Use the tasks and topics below when administering WebSphere Voice Server: v Using the administrative console Filtering on the administrative console on page 121 Enabling Security Monitoring voice servers and resource status on page 122 Starting voice servers on page 125 Stopping voice servers on page 125 v Managing using scripting on page 126 v Automatically restarting server processes on page 135 v Starting and stopping the IBM HTTP Server on page 142
Using the administrative console

The administrative console is a graphical interface for performing deployment and system administration tasks. It runs in your Web browser. You can use the console to perform tasks such as: v Start and stop voice servers v Enable product security, including access to the administrative console v Monitor voice server and resource status Starting and stopping the administrative console helps you begin using the console so that you can explore the available options.
Steps for this task

Start the server for the administrative console for a single machine topology, or start the Network Deployment manager process for multiple machine topologies. 2. Access the administrative console. After you point a Web browser at the URL for the administrative console, enter your user ID and, if needed, a password on a Login page. 3. Browse the administrative console. The important interfaces for general-purpose voice server administration are accessed using the Voice Servers panel. 4. Specify console preferences. 1.
Starting and stopping the administrative console

117
To access the administrative console, you must start it and then log in. After you finish working in the console, save your work and log out.
Steps for this task

1. Start the administrative console. a. Enable cookies in the Web browser that you use to access the administrative console in order for the administrative console to work correctly. b. In the same Web browser, type http:// your_fully_qualified_server_name:9090/admin where your_fully_qualified_server_name is the fully qualified host name for the machine that contains the administrative server. When the administrative console is on the local machine, your_fully_qualified_server_name can be localhost. If security is enabled, your request is redirected to https://your_fully_qualified_server_name:9043/ admin. c. Wait for the console to load into the browser. If you cannot start the administrative console because the console port conflicts with a Web application already running on the same machine, you must change the port number used by administrative console in the following files:
WAS_ROOT/config/cells/cell_name/nodes/node_name/servers/server_name/server.xml WAS_ROOT/config/cells/cell_name/virtualhosts.xml
Change all occurrences of port 9090 to the desired port for the console. Alternatively, shut down the other application that uses the conflicting port before starting the WebSphere Voice Server product. 2. A Login page appears after the console starts. Log into the console: a. Enter your user name (user ID). Changes made to server configurations are saved using the specified user ID. Temporary server configurations are saved to the user ID if there is a session timeout or if the user logs out without saving the configuration changes. A user ID lasts for the duration of the session for which it was used to log in. If you enter an ID that is already in use (and in session), you are prompted to do one of the following: v Force the existing user ID out of session. The configuration file used by the existing user ID is saved in the temporary area. v Wait for the existing user ID to log out or time out of the session. v Specify a different user ID. b. If the console is secure, you must also enter a password for the user name. The console is secure if the following steps have been performed: 1) Specified security user IDs and passwords 2) Enabled global security c. Click OK. 3. Stop the administrative console. Click Save on the console taskbar to save work that you have done and then click Logout to exit the console. Note that if you close the browser before saving your work, when you next log in under the same user ID, you can recover any unsaved changes.
What to do next
If you have difficulty using the administrative console on Linux, try using the Netscape Communicator 7.1 browser based on Mozilla 1.0. The browser release is
118
not officially supported by the WebSphere Voice Server product, but users have been able to access the console successfully with it. If you have difficulty using the administrative console on AIX, try using the Netscape Communicator 4.8 (or later) browser.
Login settings
Use this page to specify the user for the WebSphere Voice Server administrative console. If you are using global security, then you must also specify a password. Specifying a user allows you to resume work done previously with the product. After you type in a user ID, click OK to proceed to the next page and access the administrative console. To view this page, start the administrative console. The following are fields and settings relevant to the login window. User ID Specifies an entry identifying the user. The user ID must be unique to the administrative server. While logged in, any changes you make during the session are saved to a configuration identified by your user ID. The next time you log in to the console with your user ID, your work will be restored. If you are logging in from a client machine that currently has a browser open on the administrative console and you connect to the same server, your two connections might share the same HTTP session object, whether or not you are logging in under different user IDs. You must use two different types of browsers (for example, Microsoft Internet Explorer and Netscape Communicator) to log in to the administrative console from the same client machine in order to work from two different HTTP session objects, whether or not you use the same user ID. That is, if you need two concurrent sessions on the same client machine, access the administrative console from two different browser types. Note that you can use the same browser type to log in twice to the console, and use different user IDs. For example, suppose that Joe logs into the administrative console under the user ID joe on the physical host myhost1 using an Internet Explorer browser. Joe can log into the same console a second time using an Internet Explorer browser, but he must use a different user ID (such as peggy). Another user is currently logged in with the same user name Specifies whether to log out the user and continue work with the user ID specified or to return to the Login page and specify a different user ID or wait for the user to log out. This field appears if the user closed an administrative console browser window without logging out, then opened a new browser and tried accessing the administrative console, or if there is already another user logged in under the same name. Work with the master configuration When enabled, this setting specifies that you want to use the default administrative configuration and not use the configuration last used for the users session. Changes made to the users session since the last saving of the administrative configuration will be lost.
119
This field appears only if the user changed the administrative configuration and then logged out or closed the Web browser without saving the changes. Recover changes made in prior session When enabled, this setting specifies that you want to use the same administrative configuration last used for the users session. Recovers unsaved changes to the administrative configuration during the last users session. This field appears only if the user changed the administrative configuration and then logged out without saving the changes.
Administrative console areas and icons

Use the administrative console to create and manage resources, applications and servers, or to view product messages. To view the administrative console, ensure that the application server for the administrative console is running. Enter the URL for the administrative console into a Web browser. Then enter your user ID, and if needed, a password on the Login page. The console has the following main areas, which you can resize as desired: Taskbar The taskbar offers options for returning to the console Home page, saving changes to administrative configurations, specifying console-wide preferences, logging out of the console, and accessing product information. Navigation tree The navigation tree on the left side of the console offers links to console pages that you use to create and manage components in a WebSphere Voice Server administrative cell. Click on a + beside a tree folder or item to expand the tree for the folder or item. Click on a - to collapse the tree for the folder or item. Click on an item in the tree view to toggle its state between expanded and collapsed. Workspace The workspace on the right side of the console contains pages that you use to create and manage configuration objects such as servers and resources. Click links in the navigation tree to view the different types of configured objects. Within the workspace, click on configured objects to view their configurations, run-time status, and options. WebSphere Status The status messages area at the bottom of the console provides information on messages returned by the WebSphere Voice Server. The information deals with problems in your administrative configuration as well as messages about run-time events. Click Previous or Next to toggle among displays of configuration problems and run-time events. You can adjust the interval between automatic refreshes in the Preferences settings. For details about administrative console buttons, page features, console navigation, and taskbar actions, refer to the WebSphere Application Server Network Deployment 5.1.x Information Center using the links below. Note that not all the
120
console features mentioned in the WebSphere Application Server information center are available in WebSphere Voice Server. v Information on buttons v Page features v Console navigation v Taskbar actions For details specific to the WebSphere Voice Server administrative console, see Monitoring voice servers and resource status
Administrative console icons

The administrative console uses the following icons to indicate voice server status.
Icon Meaning Indicates a partial start of the server Indicates a partial stop of the server Indicates that the server is running Indicates that the server is stopped Indicates that the server status is unavailable Indicates that the server status is unknown
Filtering on the administrative console

On the Voice Servers panel, there is a Filter option which allows you to filter the information by columns in the Overall Status table to best suit your needs. Use these steps to filter on the administrative console: 1. Expand the Filter options by clicking the plus (+). 2. Using the drop-down menu, select the column by which to filter. The options are Voice Server Name and Status. Note: The filter is case sensitive, so make sure that when you enter a filter value for Voice Server Name or Status, the cases you type match those used by the console. For example, the following Status values are accepted: v Started v Stopped v Partial Start v Unavailable v Unknown 3. Enter any filter criteria and click Go. The wild card characters are: *, ?, % For example, if you are filtering on Voice Server Name and you know that the machines you want to check all have the term server as part of their names,
121
type *server* in the filter field and click Go. The name and status of each machine with server in the name are displayed the Overall Status table.
Monitoring voice servers and resource status

The administrative console has specific panels for WebSphere Voice Server information.
Voice Servers panel

The Voice Servers panel is used to monitor the status of your voice servers. It is the default view when you log in to the administrative console. There are two tables on the main Voice Servers panel, the Overall Status table and the WebSphere Status table. Overall Status table By viewing the Overall Status table, you can monitor the status of your entire voice server system, and you can filter the information in this table to suit your needs. To start or stop a voice server, check the box beside the server you want to start or stop. Then click the Start or Stop button. The following items are available for monitoring in the Overall Status table. Overall Status An icon depicts the overall status of your voice servers. v If the automatic speech recognition (ASR) and text-to-speech (TTS) engines are started, the voice server is started. v If the ASR and/or TTS engines are partially started, the voice server is partially started. v If either the ASR or TTS engines are stopped and the other is not stopped, the voice server is partially started. v If both the ASR and TTS engines are stopped, the voice server is stopped. Total The number of voice servers in the current system.
Voice Server Name This column displays the names of your voice servers using the convention node name_server1. Click a server link, such as wvssvr1_server1, to display detailed configuration and runtime status. Status An icon depicts the status of your individual voice servers. A Refresh icon updates the status icons for all voice servers when clicked. WebSphere Status Use the WebSphere Status area of the administrative console to view error and runtime event messages returned by WebSphere Voice Server. The WebSphere Status area displays along the bottom of the console and remains visible as you navigate from the Voice Servers panel to other pages.
122
The area displays two frames: WebSphere Configuration Problems and WebSphere Runtime Messages. Click Previous or Next to toggle between the frames. Click the Refresh icon in the upper-right of the area to refresh the information displayed. You can adjust the interval between automatic refreshes in the Preferences settings. WebSphere Configuration Problems Displays the number of workspace files. This frame also displays the number of problems with the administrative configuration for the user ID. Click the number to view detailed information on the problems. WebSphere Runtime Message Displays the number of messages returned by WebSphere Voice Server as well as the number of error messages (x icon), warning messages (! icon), and informational messages (i icon). Click the number of messages to view details.
Configuration and runtime settings for specific voice servers

When you click a voice server link, such as wvssvr1_server1, in the Voice Servers panel, a new view displays detailed configuration and runtime status for that specific server. You can toggle between the Configuration and Runtime panels using the appropriately labeled tabs. Clicking any Back button will return you to the Voice Servers panel. Configuration tab This view lets you monitor and configure a specific voice server. ASR Configuration This table shows current configuration settings and controls for ASR resources. For details about this table, refer to Configuring ASR settings. TTS Configuration This table shows current configuration settings and controls for TTS resources. For more details, refer to Configuring TTS settings. Runtime tab This view shows the runtime status for a specific server. It is the visible or default interface when you click a specific voice server link. General Status This table provides general resource status for CPU and memory utilization, and it provides the overall status of the WebSphere Voice Server supporting components. The first column shows the status item name, the second column provides a graphical display of the current setting, and the third column presents a short description of the status item Performance Monitoring Infrastructure (PMI) WebSphere Voice Server collects run time data through the Performance Monitoring Infrastructure (PMI). PMI is automatically enabled for a
123
single-server environment when you install WebSphere Voice Server. To enable PMI services in WebSphere Application Server Network Deployment, see the topics below: v Single-server environment (enabled automatically when you install WebSphere Voice Server): Enabling performance monitoring services in the application server through the administrative console. v Mutliple-server environment (cell): Enabling performance monitoring services in the Node Agent through the administrative console . ASR Status This table provides general resource status for ASR resources. The first column shows the status item name, the second column provides a graphical or text display of the current status, and the third column presents a short description of the status item. The three status items are ASR Engines, Active Sessions, and Languages installed. The ASR engine status is shown using an icon. Click the ASR engine status icon to display the status name. v If the ASR engine status icon is a filled-in green arrow ( component is started. ), the
v If the ASR engine status icon is a white arrow outlined in green ( the component is partially started. v If the ASR engine status is a red X ( ), the component is stopped.
),
If the number of active ASR sessions is less than the total, then the WebSphere Voice Server is not running to its full configuration capacity. If this number is always less than the total, then the total number of ASR engines could be reduced to save system resources. If this number is always equal to the total, check the logs to make sure a sufficient number of ASR engines are configured. You might need to increase the total number of configured ASR engines. TTS Status This table provides general resource status for TTS resources. The first column shows the status item name, the second column provides a graphical or text display of the current status, and the third column presents a short description of the status item. The four status items are TTS Engines, Active Sessions, Default voice, and Voice(s) installed. TTS voices that are available for use are marked with an asterisk (*) on the administrative console. For details about the differences between installed voices and available voices for TTS, refer to Configuring default and available TTS voices on page 104. TTS engine status is shown using an icon. Click the TTS engine status icon to display the status name. v If the TTS engine status icon is a filled-in green arrow ( component is started. ), the
v If the TTS engine status icon is a white arrow outlined in green ( the component is partially started. v If the TTS engine status is a red X ( ), the component is stopped.
),
If the number of active TTS sessions is less than the total, then the WebSphere Voice Server is not running to its full configuration capacity. If
124
this number is always less than the total, then that the total number of TTS engines could be reduced to save system resources. If this number is always equal to the total, check the logs to make sure a sufficient number of TTS engines are configured. You might need to increase the total number of configured TTS engines.
Starting voice servers

Voice servers can be started using the administrative console or scripting. Use the following steps to start a voice server in the administrative console: 1. In the Voice Servers panel, identify the voice server you would like to start (note the Status icon to determine the current status of the voice server) and click in the check box to the left of Voice Server Name. 2. Click the Start button. The status icon should appear as a green arrow pointing to the right ( ) when the voice server has started.
Alternatively, you can use the startVoiceServer wsadmin sample to start a single voice server from the command line using wsadmin.
Known issue starting servers on a Windows Server 2003 platform

Occasionally an issue regarding process inheritance of socket handles on Windows platforms can result in an over-allocation of system handles, reducing available system resources. The situation occurs after repeated starting and stopping of the WebSphere Voice Server Enterprise Application Archive (EAR). To avoid this situation, restart WebSphere Application Server from the command line using the stopServer and startServer commands when restarting the WebSphere Voice Server application EAR.
Stopping voice servers

Voice servers can be stopped using the administrative console or scripting. Use the following steps to stop a server in the administrative console: 1. In the Voice Servers panel, identify the voice server you would like to stop (note the Status icon to determine the current status of the voice server) and click in the check box to the left of Voice Server Name. 2. Click the Stop button. The status icon should appear as a red X ( voice server has stopped. ) when the
Alternatively, you can use the stopVoiceServer wsadmin sample to stop a single voice server from the command line using wsadmin.
Security
IBM WebSphere Voice Server uses the security infrastructure and mechanisms provided by WebSphere Application Server. WebSphere Application Server allows administrators to protect a server using two security features: Global Security and Java 2 Security. These security features are managed through the Security panel on the administrative console.
125
Important: Java 2 Security cannot be enabled in order for WebSphere Voice Server to function properly. If you need security, enable Global Security in the WebSphere Voice Server administrative console. Do not select the option to enforce Java 2 Security. Before you enable global security read the technote about using Local operating system user registries. For detailed information about the WebSphere approach to security, refer to Welcome to Security in the WebSphere Application Server Information center.
Managing using scripting

WebSphere Voice Server provides sample wsadmin scripts for use with a single node installation. Important: These scripts are intended to teach new WebSphere Voice Server administrators about potential uses of scripting and are not appropriate for use in production systems. The WebSphere Application Server wsadmin tool provides the ability to run scripts. You can use the wsadmin tool to manage a WebSphere Voice Server installation, configuration, application deployment, and server runtime operations. The WebSphere Application Server only supports Jacl, an alternate implementation of the interpreted script language Tcl (Tool Command Language), written in Java. Note: Basic knowledge of the syntax for the Jacl scripting language is helpful in order to understand and modify the samples. For more information about Jacl, refer to Scripting: resources for learning. These scripts are provided as samples to allow convenient access to common WebSphere Voice Server administrator tasks; starting and stopping voice servers, obtaining the voice server status, and backing up and restoring voice server configurations. You must first issue the WebSphere Application Server startServer command, and then issue the startVoiceServer command. Because each production environment is unique, these samples provide templates to assist new users in using wsadmin and Jacl. It is assumed that administrators will modify the code provided for their production environment (such as enabling WebSphere Application Server global security). For more information on scripting, see Deploying and managing using scripting.
Sample scripts
The sample scripts assist with common administrator tasks that can be performed using wsadmin scripts. These samples are provided for convenience and to serve as examples. They include the following: v getParam sample on page 127 v putParam sample on page 128 v startVoiceServer sample on page 129 v stopVoiceServer sample on page 130 v statusVoiceServer sample on page 131
126
v backupVoiceServerConfig sample on page 132 v restoreVoiceServerConfig sample on page 134 These sample scripts are installed into and are available from the samples directory in the root of your WebSphere Voice Server installation directory.
getParam sample
The getParam sample queries and displays WebSphere Voice Server configuration parameters from any node (or nodes) in the cell. Note: WebSphere Application Server must be started in order for this script to work correctly. If WebSphere Voice Server has global security enabled, then sample utilities must be updated to use name and password parameters. To use this sample on AIX or Linux, issue the following:
cd $WVS_ROOT/samples ./getParam.sh
To use this sample on Windows, issue the following:

cd $WVS_ROOT\samples getParam.bat
Syntax
The samples syntax is as follows:
getParam parameter scope
parameter This variable is the parameter for which you are requesting information. A terminal wild card is allowed in a parameter name. Use the dollar ($) character to denote a wild card, such as com.ibm.$. scope This variable is the host name of the target. A terminal wild card is allowed in a parameter name. Use the dollar ($) character to denote a wild card, such as wvssrv$. Note: The scope is optional if you are not using Deployment Manager. The target defaults to localhost if no other value is specified. If you are using a Deployment Manager environment, you must define the scope.
Examples
getParam com.ibm.voice.server.synth.mrcp.Andrew.$ getParam com.ibm.voice.server.synth.mrcp.Andrew.$ ttssrv3 getParam com.ibm.voice.server.synth.mrcp.Andrew.$ ttssrv$
Other sample scripts putParam sample The putParam sample sets WebSphere Voice Server configuration parameters for any node (or nodes) in the cell. startVoiceServer sample The startVoiceServer sample starts WebSphere Application Server and WebSphere Voice Server.
127
stopVoiceServer sample The stopVoiceServer sample stops the WebSphere Voice Server Enterprise Archive (EAR). statusVoiceServer sample The statusVoiceServer sample obtains the current status of a voice server. backupVoiceServerConfig sample The backupVoiceServerConfig sample script backs up the WebSphere Voice Server configuration and saves it to a zipped file. This sample should only be run on a machine where the WebSphere Voice Server has been installed. restoreVoiceServerConfig sample The restoreVoiceServerConfig sample restores the WebSphere Voice Server configuration captured in a previous backupVoiceServerConfig for the same machine. You should only run this sample on a machine where WebSphere Voice Server has been installed.
putParam sample
The putParam sample sets WebSphere Voice Server configuration parameters for any node (or nodes) in the cell. Note: WebSphere Application Server must be started in order for this script to work correctly. If WebSphere Voice Server has global security enabled, then sample utilities must be updated to use name and password parameters. To use this sample on AIX or Linux, issue the following:
cd $WVS_ROOT/samples ./putParam.sh

cd $WVS_ROOT\samples putParam.bat
Syntax
putParam parameter value scope
parameter This variable is the parameter you want to set. value This variable is the value assigned to a specified parameter. Note: To set this variable to spaces ( ), use the value of null or NULL. scope This variable is the host name of the target. A terminal wild card is allowed in a parameter name. Use the dollar ($) character to denote a wild card, such as wvssrv$. Note: The scope is optional if you are not using Deployment Manager. The target defaults to localhost if no other value is specified. If you are using a Deployment Manager environment, you must define the scope.
128
Examples
putParam putParam putParam putParam com.ibm.voice.server.synth.pool-size 10 com.ibm.voice.server.synth.pool-size 10 ttssrv3 com.ibm.voice.server.synth.pool-size 10 ttssrv$ com.ibm.voice.server.recognizer.grammar.preload null
ttssrv$
Other sample scripts getParam sample The getParam sample queries and displays WebSphere Voice Server configuration parameters from any node (or nodes) in the cell. startVoiceServer sample The startVoiceServer sample starts WebSphere Application Server and WebSphere Voice Server. stopVoiceServer sample The stopVoiceServer sample stops the WebSphere Voice Server Enterprise Archive (EAR). statusVoiceServer sample The statusVoiceServer sample obtains the current status of a voice server. backupVoiceServerConfig sample The backupVoiceServerConfig sample script backs up the WebSphere Voice Server configuration and saves it to a zipped file. This sample should only be run on a machine where the WebSphere Voice Server has been installed. restoreVoiceServerConfig sample The restoreVoiceServerConfig sample restores the WebSphere Voice Server configuration captured in a previous backupVoiceServerConfig for the same machine. You should only run this sample on a machine where WebSphere Voice Server has been installed.
startVoiceServer sample
The startVoiceServer sample starts WebSphere Application Server and WebSphere Voice Server. On a base WebSphere Voice Server without Network Deployment, you must use the WebSphere Application Server startServer command line utility before running the startVoiceServer.jacl script. Use the startServer command as follows: To use this sample on AIX or Linux, issue the following:
cd WAS_ROOT/bin ./startServer.sh server1
On Windows:
cd WAS_ROOT\bin startServer server1
WAS_ROOT is where WebSphere Application Server is installed and server1 is the name of your default WebSphere Application Server. If WebSphere Voice Server has global security enabled, then sample utilities must be updated to use name and password parameters. To use this sample on AIX or Linux, issue the following:
cd $WVS_ROOT/samples ./startVoiceServer.sh

129
cd $WVS_ROOT\samples startVoiceServer
The startVoiceServer script performs either one or both of the following actions: v Starts the default WebSphere Application Server, server1. v If the WebSphere Voice Server EAR is not started, this utility will start it.
Syntax
startVoiceServer [parameter]
Parameters
The following parameter is available for the startVoiceServer sample: --help Prints a usage statement. Other sample scripts getParam sample The getParam sample queries and displays WebSphere Voice Server configuration parameters from any node (or nodes) in the cell. putParam sample The putParam sample sets WebSphere Voice Server configuration parameters for any node (or nodes) in the cell. stopVoiceServer sample The stopVoiceServer sample stops the WebSphere Voice Server Enterprise Archive (EAR). statusVoiceServer sample The statusVoiceServer sample obtains the current status of a voice server. backupVoiceServerConfig sample The backupVoiceServerConfig sample script backs up the WebSphere Voice Server configuration and saves it to a zipped file. This sample should only be run on a machine where the WebSphere Voice Server has been installed. restoreVoiceServerConfig sample The restoreVoiceServerConfig sample restores the WebSphere Voice Server configuration captured in a previous backupVoiceServerConfig for the same machine. You should only run this sample on a machine where WebSphere Voice Server has been installed.
stopVoiceServer sample
The stopVoiceServer sample stops the WebSphere Voice Server Enterprise Archive (EAR). If WebSphere Application Server has global security enabled, then sample utilities must be updated to use name and password parameters. To use this sample on AIX or Linux, issue the following:
cd $WVS_ROOT/samples ./stopVoiceServer.sh

cd $WVS_ROOT\samples stopVoiceServer
130
This sample uses the stopVoiceServer.jacl script to perform the following actions: v If the WebSphere Voice Server EAR is started, this script will stop it. v If the -wvsonly parameter is not specified, then the default WebSphere Application Server will be stopped. For information on the WebSphere Application Server stopServer command, see the WebSphere Application Server Network Deployment 5.1.x Information Center.
Syntax
The sample syntax is as follows:
stopVoiceServer [parameter]
Parameters
The following parameters are available for the stopVoiceServer sample: -wvsonly Stops only the WebSphere Voice Server EAR. --help Prints a usage statement. Other sample scripts getParam sample The getParam sample queries and displays WebSphere Voice Server configuration parameters from any node (or nodes) in the cell. putParam sample The putParam sample sets WebSphere Voice Server configuration parameters for any node (or nodes) in the cell. startVoiceServer sample The startVoiceServer sample starts WebSphere Application Server and WebSphere Voice Server. statusVoiceServer sample The statusVoiceServer sample obtains the current status of a voice server. backupVoiceServerConfig sample The backupVoiceServerConfig sample script backs up the WebSphere Voice Server configuration and saves it to a zipped file. This sample should only be run on a machine where the WebSphere Voice Server has been installed. restoreVoiceServerConfig sample The restoreVoiceServerConfig sample restores the WebSphere Voice Server configuration captured in a previous backupVoiceServerConfig for the same machine. You should only run this sample on a machine where WebSphere Voice Server has been installed.
statusVoiceServer sample
The statusVoiceServer sample obtains the current status of a voice server. If WebSphere Application Server has global security enabled, then sample utilities must be updated to use name and password parameters. To use this sample on AIX or Linux, issue the following:
cd $WVS_ROOT/samples ./statusVoiceServer.sh
131

cd $WVS_ROOT\samples statusVoiceServer
This sample uses the statusVoiceServer.jacl script to perform the following actions: v Queries the status of the WebSphere Voice Server Enterprise Archive (EAR) application, WVS5.1-node. If the EAR is installed and started, and if the managed bean (MBean), WVSAdminControl is not found, the voice server is stopped. v If WebSphere Voice Server EAR is started, then the WVSAdminControl MBean getStatus operation is queried for additional runtime status such as Started, Stopped, Partial Start.
Syntax
statusVoiceServer [parameter]
Parameters
The following parameter is available for the statusVoiceServer sample: --help Prints a usage statement. Other sample scripts getParam sample The getParam sample queries and displays WebSphere Voice Server configuration parameters from any node (or nodes) in the cell. putParam sample The putParam sample sets WebSphere Voice Server configuration parameters for any node (or nodes) in the cell. startVoiceServer sample The startVoiceServer sample starts WebSphere Application Server and WebSphere Voice Server. stopVoiceServer sample The stopVoiceServer sample stops the WebSphere Voice Server Enterprise Archive (EAR). backupVoiceServerConfig sample The backupVoiceServerConfig sample script backs up the WebSphere Voice Server configuration and saves it to a zipped file. This sample should only be run on a machine where the WebSphere Voice Server has been installed. restoreVoiceServerConfig sample The restoreVoiceServerConfig sample restores the WebSphere Voice Server configuration captured in a previous backupVoiceServerConfig for the same machine. You should only run this sample on a machine where WebSphere Voice Server has been installed.
backupVoiceServerConfig sample
The backupVoiceServerConfig sample script backs up the WebSphere Voice Server configuration and saves it to a zipped file. This sample should only be run on a machine where the WebSphere Voice Server has been installed. If WebSphere Application Server has global security enabled, then sample utilities must be updated to use name and password parameters.
132
To use this sample on AIX and Linux, issue the following:

cd $WVS_ROOT/samples ./backupVoiceServerConfig.sh

cd $WVS_ROOT\samples backupVoiceServerConfig
This sample uses the backupVoiceServerConfig.jacl script to perform the following actions: v The WebSphere Voice Server Enterprise Archive (EAR) will be stopped. This ensures that the configuration documents are not locked by WebSphere Voice Server and will not change while it is being backed up. You will have to restart WebSphere Voice Server EAR after the backup is complete. v If the backup_file parameter (where backup_file is a name you have selected) is not specified, then the script creates a file in the current directory using the following name (overwriting the file if it exists already): WVSConfig_yyyy-mm-dd.zip where yyyy is the current year, mm is the current month and dd is the current day v Uses the WVSAdminConfig managed bean (MBean) backup operation to store the Voice Server configuration into the backup_file (zipped file format). The backupVoiceServerConfig sample does not capture the product binaries or capture application data stored elsewhere (including on an HTTP Server). As such, the backupVoiceServerConfig utility is not a substitute for a comprehensive system backup facility.
Syntax
backupVoiceServerConfig [option]
Note: If you are backing up a server, you must specify the node name.
Parameters
The following parameters are available for the backupVoiceServerConfig sample: backup_file An optional variable that specifies the location of the backup file to which information gets written. --help Prints a usage statement. Other sample scripts getParam sample The getParam sample queries and displays WebSphere Voice Server configuration parameters from any node (or nodes) in the cell. putParam sample The putParam sample sets WebSphere Voice Server configuration parameters for any node (or nodes) in the cell. startVoiceServer sample The startVoiceServer sample starts WebSphere Application Server and WebSphere Voice Server.
133
stopVoiceServer sample The stopVoiceServer sample stops the WebSphere Voice Server Enterprise Archive (EAR). statusVoiceServer sample The statusVoiceServer sample obtains the current status of a voice server. restoreVoiceServerConfig sample The restoreVoiceServerConfig sample restores the WebSphere Voice Server configuration captured in a previous backupVoiceServerConfig for the same machine. You should only run this sample on a machine where WebSphere Voice Server has been installed.
restoreVoiceServerConfig sample
The restoreVoiceServerConfig sample restores the WebSphere Voice Server configuration captured in a previous backupVoiceServerConfig for the same machine. You should only run this sample on a machine where WebSphere Voice Server has been installed. If WebSphere Application Server has global security enabled, then sample utilities must be updated to use name and password parameters. To use this sample on AIX or Linux, issue the following:
cd $WVS_ROOT/samples ./restoreVoiceServerConfig.sh

cd $WVS_ROOT\samples restoreVoiceServerConfig
You must specify the name of the zip file generated by thebackupVoiceServerConfig utility. This sample uses the restoreVoiceServerConfig.jacl script to perform the following actions: v The WebSphere Voice Server Enterprise Archive (EAR) will be stopped. This ensures there are no sharing or read conflicts between the Voice Server and the restore processing. You will have to restart the Voice Server EAR after the restore is complete. v Uses the WVSAdminConfig managed bean (MBean) restore operation to restore the WebSphere Voice Server configuration from the backup_file (zip file format), overwriting the current configuration.
Syntax
restoreVoiceServerConfig backup_file [option]
Parameters
The following parameters are available for the restoreVoiceServerConfig sample: backup_file A required file name that specifies the location of the backup file to use for restoring the configuration.
134
--help Prints a usage statement. Other sample scripts getParam sample The getParam sample queries and displays WebSphere Voice Server configuration parameters from any node (or nodes) in the cell. putParam sample The putParam sample sets WebSphere Voice Server configuration parameters for any node (or nodes) in the cell. startVoiceServer sample The startVoiceServer sample starts WebSphere Application Server and WebSphere Voice Server. stopVoiceServer sample The stopVoiceServer sample stops the WebSphere Voice Server Enterprise Archive (EAR). statusVoiceServer sample The statusVoiceServer sample obtains the current status of a voice server. backupVoiceServerConfig sample The backupVoiceServerConfig sample script backs up the WebSphere Voice Server configuration and saves it to a zipped file. This sample should only be run on a machine where the WebSphere Voice Server has been installed.
Automatically restarting server processes

There are several server processes related to WebSphere Voice Server that the operating system can monitor and automatically restart when the server processes stop abnormally. This task describes how to update the operating system configuration to automatically restart WebSphere Voice Server during a system startup.
Before you begin

Once your system is validated, you may choose to configure the system to automatically start WebSphere Voice Server during system startup.WebSphere Voice Server starts when WebSphere Application Server is started. If you set up WebSphere Application Server to start automatically, then WebSphere Voice Server will also start automatically. For more details, see Automatically restarting server processes in the WebSphere Application Server Information Center. v Restarting on AIX platforms v Restarting on Red Hat Linux platforms on page 136 v Restarting on SUSE LINUX platforms on page 137 v Restarting on Windows Server 2003 platforms on page 138
Restarting on AIX platforms

AIX provides a standardized interface to allow users to add scripts to start various processes during system initialization without requiring a user to log in to the system. This process consists of three steps: 1. Create the start_server1 script by navigating to the /usr/WebSphere/ AppServer/bin directory and running the following command: ./startServer.sh server1 -script -background This command generates a script named start_server1.sh which is referenced by the actual initialization script.
135
2. Create the ibmhttpd and was scripts using the sample scripts below for reference and place them in the /etc/rc.d/rc5.d directory. Make sure to set the executable flag in the scripts. The ibmhttpd script is the system initialization script that is used to automatically start the IBM HTTP Server at system initialization time. The was script is the system initialization script that is used to automatically start WebSphere Application Server and WebSphere Voice Server. Each script should have options for starting and stopping the desired modules. Important: Be sure to validate the order in which the start and stop script links are run in the rc5.d directory. These links all have a Snn or Knn designation, where the nn is a numerical value which determines the start order for those scripts. The Snnibmhttpd and Snnwas scripts must have a value which will enable them to start after the network has been started. Similarly, the Knn scripts (if they exist) must be run before the network stop script. For example, the following four files would be in your rc5.d directory:
S1_ibmhttpd.sh S2_was.sh K1_was.sh K2_ibmhttpd.sh
At startup, the S1_ibmhttpd.sh script runs first, and then the S2_was.sh runs. During shutdown, the K1_was.sh script is called first, and then the K2_ibmhttpd.sh script. 3. Test the autostart scripts by rebooting your system and verifying that the desired processes have started.
Restarting on Red Hat Linux platforms

Red Hat Linux provides a standardized interface to allow users to add scripts to start various processes during system initialization without requiring a user to log in to the system. This process consists of three steps: 1. Create the start_server1 script by navigating to the /opt/WebSphere/ AppServer/bin directory and running the following command: ./startServer.sh server1 -script -background This command generates a script named start_server1.sh which is referenced by the actual initialization script. 2. Create the ibmhttpd and was scripts using the sample scripts below for reference and place them in the /etc/init.d directory. Make sure to set the executable flag in the scripts. The ibmhttpd script is the system initialization script that is used to automatically start the IBM HTTP Server at system initialization time. The was script is the system initialization script that is used to automatically start WebSphere Application Server and WebSphere Voice Server. Each script should have options for starting and stopping the desired modules. 3. Establish the ibmhttpd and was scripts as services in order to run them in the system initialization process. To do this, enter the following commands as root user:
chkconfig chkconfig chkconfig chkconfig --add was --level 5 was on --add ibmhttpd --level 5 ibmhttpd
136
Important: Each of these commands has two of the hyphen characters repeated. A single hyphen will not work when using the full text name of the tag. The chkconfig --add commands add the script entries into the services table and the chkconfig --level commands indicate the runlevels at which Red Hat should automatically run the scripts. 4. Test the autostart scripts by rebooting your system and verifying that the desired processes have started.
Restarting on SUSE LINUX platforms

SUSE LINUX provides a standardized interface to allow users to add scripts to start various processes during system initialization without requiring a user to log in to the system. This process consists of three steps: 1. Create the start_server1 script by navigating to the /opt/WebSphere/ AppServer/bin directory and running the following command: ./startServer.sh server1 -script -background This command generates a script named start_server1.sh which is referenced by the actual initialization script. 2. Create the ibmhttpd and was scripts using the sample scripts below for reference and place them in the /etc/init.d directory. Make sure to set the executable flag in the scripts. The ibmhttpd script is the system initialization script that is used to automatically start the IBM HTTP Server at system initialization time. The was script is the system initialization script that is used to automatically start WebSphere Application Server and WebSphere Voice Server. Each script should have options for starting and stopping the desired modules. 3. Establish the ibmhttpd and was scripts as services in order to run them in the system initialization process. To do this, enter the following commands as root user:
chkconfig chkconfig chkconfig chkconfig --add --set --add --set was was 5 ibmhttpd ibmhttpd 5
Important: Each of these commands has two of the hyphen characters repeated. A single hyphen will not work when using the full text name of the tag. The chkconfig --add commands add the script entries into the services table and the chkconfig --set commands indicate the runlevels at which SUSE should automatically run the scripts. The chkconfig --set commands create links in the rc5.d directory which are run at system initialization and shutdown. These links all have a Snn or Knn designation, where the nn is a numerical value which determines the start order for those scripts. Important: Each of these commands have two of the hyphen characters repeated. A single hyphen will not work when using the full text name of the tag. Also, be sure to validate the order in which the start and stop script links are run in the rc5.d directory. The Snnibmhttpd and Snnwas scripts must have a value which will enable them to start after the network has been started. Similarly, the Knn scripts (if they exist) must be run before the network stop script. For example, the following four files would be in your rc5.d directory:
137
S1_ibmhttpd.sh S2_was.sh K1_was.sh K2_ibmhttpd.sh
At startup, the S1_ibmhttpd.sh script runs first, and then the S2_was.sh runs. During shutdown, the K1_was.sh script is called first, and then the K2_ibmhttpd.sh script. 4. Test the autostart scripts by rebooting your system and verifying that the desired processes have started.
Restarting on Windows Server 2003 platforms

When you installed WebSphere Voice Server you were given the option of asking the installation wizard to create services entries for both IBM HTTP Server server as well as WebSphere Application Server. If you selected both to be generated, you only need to edit the properties of those services to request that the services be automatically started during system initialization without requiring a user to log in to the system. By default, the IBM HTTP Server is set to start automatically. To complete the automatic start process, you need to update the setting for the WebSphere Application Server Administrative Server.
Sample scripts for AIX

Example of ibmhttpd script:
#!/bin/bash # # apache # # description: Start up the Apache web server. RETVAL=$? APACHE_HOME="/usr/IBMHttpServer" case "$1" in start) if [ -f $APACHE_HOME/bin/apachectl ]; then echo $"Starting IBM Http Server" $APACHE_HOME/bin/apachectl start fi ;; stop) if [ -f $APACHE_HOME/bin/apachectl ]; then echo $"Stopping IBM Http Server" $APACHE_HOME/bin/apachectl stop fi ;; status) if [ -f $APACHE_HOME/bin/apachectl ]; then echo $"Show status of IBM Http Server" $APACHE_HOME/bin/apachectl status fi ;; *) echo $"Usage: $0 {start|stop|status}" exit 1 ;; esac exit $RETVAL
138
For a standalone WebSphere Voice Server system, the start script only needs to start the default server1. The sample script described is listed in the following example. Example of a was autostart script:
#!/bin/bash # # apache # # description: Start up the WebSphere Application Server. RETVAL=$? WAS_HOME="/usr/WebSphere/AppServer" case "$1" in start) if [ -f $WAS_HOME/bin/start_server1.sh ]; then echo $"Starting IBM WebSphere Application Server" $WAS_HOME/bin/start_server1.sh fi ;; stop) if [ -f $WAS_HOME/bin/stopServer.sh ]; then echo $"Stop IBM WebSphere Application Server" $WAS_HOME/bin/stopServer.sh server1 fi ;; status) if [ -f $WAS_HOME/bin/serverStatus.sh ]; then echo $"Show status of IBM WebSphere Application Server" $WAS_HOME/bin/serverStatus.sh server1 fi ;; *) echo $"Usage: $0 {start|stop|status}" exit 1 ;; esac exit $RETVAL
Once you have federated your WebSphere Voice Server into a deployment manager cell, you need to start the node agent as well as the default server1. Note that this script also issues the iptables command which is needed by the WebSphere Edge Component Load Balancer. If you do not use Load Balancer, be sure to remove that line. Example of a wasNode script:
#!/bin/bash # # apache # # description: Start up the WebSphere Application Server. RETVAL=$? WAS_HOME="/usr/WebSphere/AppServer" # added line to ensure that environment variables are set correctly . /usr/WebSphere/VoiceServer/wvs.setenv case "$1" in start) if [ -f $WAS_HOME/bin/start_server1.sh ]; then echo $"Starting IBM WebSphere Node Agent and Application Server" $WAS_HOME/bin/start_nodeagent.sh $WAS_HOME/bin/start_server1.sh # added line to set iptables entry for Edge Load Balancer support iptables -t nat -A PREROUTING -d 9.42.171.99 -j REDIRECT fi ;;
139
stop) if [ -f $WAS_HOME/bin/stopServer.sh ]; then echo $"Stop IBM WebSphere Application Server" $WAS_HOME/bin/stopServer.sh server1 $WAS_HOME/bin/stopNode.sh fi ;; status) if [ -f $WAS_HOME/bin/serverStatus.sh ]; then echo $"Show status of IBM WebSphere Application Server" $WAS_HOME/bin/serverStatus.sh server1 fi ;; *) echo $"Usage: $0 {start|stop|status}" exit 1 ;; esac exit $RETVAL
Sample scripts for Linux

Example of ibmhttpd script:
#!/bin/bash # # apache # # chkconfig: 5 85 15 # description: Start up the Apache web server. RETVAL=$? APACHE_HOME="/opt/IBMHttpServer" case "$1" in start) if [ -f $APACHE_HOME/bin/apachectl ]; then echo $"Starting IBM Http Server" $APACHE_HOME/bin/apachectl start fi ;; stop) if [ -f $APACHE_HOME/bin/apachectl ]; then echo $"Stopping IBM Http Server" $APACHE_HOME/bin/apachectl stop fi ;; status) if [ -f $APACHE_HOME/bin/apachectl ]; then echo $"Show status of IBM Http Server" $APACHE_HOME/bin/apachectl status fi ;; *) echo $"Usage: $0 {start|stop|status}" exit 1 ;; esac exit $RETVAL
For a standalone WebSphere Voice Server system, the start script only needs to start the default server1. The sample script described is listed in the following example. Example of a was autostart script:
140
#!/bin/bash # # apache # # chkconfig: 5 90 10 # description: Start up the WebSphere Application Server. RETVAL=$? WAS_HOME="/opt/WebSphere/AppServer" case "$1" in start) if [ -f $WAS_HOME/bin/start_server1.sh ]; then echo $"Starting IBM WebSphere Application Server" $WAS_HOME/bin/start_server1.sh fi ;; stop) if [ -f $WAS_HOME/bin/stopServer.sh ]; then echo $"Stop IBM WebSphere Application Server" $WAS_HOME/bin/stopServer.sh server1 fi ;; status) if [ -f $WAS_HOME/bin/serverStatus.sh ]; then echo $"Show status of IBM WebSphere Application Server" $WAS_HOME/bin/serverStatus.sh server1 fi ;; *) echo $"Usage: $0 {start|stop|status}" exit 1 ;; esac exit $RETVAL
Once you have federated your WebSphere Voice Server into a deployment manager cell, you need to start the node agent as well as the default server1. Note that this script also issues the iptables command which is needed by the WebSphere Edge Component Load Balancer. If you do not use Load Balancer, be sure to remove that line. Example of a wasNode script:
#!/bin/bash # # apache # # chkconfig: 5 90 10 # description: Start up the WebSphere Application Server. RETVAL=$? WAS_HOME="/opt/WebSphere/AppServer" # added line to ensure that environment variables are set correctly . /opt/WebSphere/VoiceServer/wvs.setenv case "$1" in start) if [ -f $WAS_HOME/bin/start_server1.sh ]; then echo $"Starting IBM WebSphere Node Agent and Application Server" $WAS_HOME/bin/start_nodeagent.sh $WAS_HOME/bin/start_server1.sh # added line to set iptables entry for Edge Load Balancer support iptables -t nat -A PREROUTING -d 9.42.171.99 -j REDIRECT fi ;; stop) if [ -f $WAS_HOME/bin/stopServer.sh ]; then echo $"Stop IBM WebSphere Application Server" $WAS_HOME/bin/stopServer.sh server1
141
$WAS_HOME/bin/stopNode.sh fi ;; status) if [ -f $WAS_HOME/bin/serverStatus.sh ]; then echo $"Show status of IBM WebSphere Application Server" $WAS_HOME/bin/serverStatus.sh server1 fi ;; *) echo $"Usage: $0 {start|stop|status}" exit 1 ;; esac exit $RETVAL
Starting and stopping the IBM HTTP Server

The instructions for starting and stopping the IBM HTTP Server are dependent upon the operating system you use. This topic includes specific instructions for AIX, Linux, and Windows Server 2003 Enterprise and Standard editions.
On an AIX or Linux platform

The apachectl command is used to start and stop IBM HTTP Server. The apachectl command is located in the bin subdirectory within the IBM HTTP Server installation directory. If that directory is not in your PATH, the full path should be given in the command line. The samples below assume that the IBM HTTP Server was installed in the /usr/IBMHttpServer directory on AIX and in the /opt/IBMHttpServer directory on Linux. Use the following command to start the IBM HTTP Server: #/usr/IBMHttpServer/bin/apachectl start #/opt/IBMHttpServer/bin/apachectl start Confirm that IBM HTTP Server started successfully by opening a browser and typing your server name in the address box. Use the following command to stop the IBM HTTP Server: #/usr/IBMHttpServer/bin/apachectl stop #/opt/IBMHttpServer/bin/apachectl stop
On a Windows platform
Accept the default WebSphere Voice Server configuration, and install IBM HTTP Server Apache as a Windows service, so that IBM HTTP Server starts automatically with your machine, and keeps running after you log off. Once Apache for Windows is configured as a Windows Service, follow the samples below to manage starting and stopping IBM HTTP Server. Remember: Replace the version of the service name with the one appropriate for your installation.
142
Issue the following command from a console window to start the IBM HTTP Server: NET START IBM HTTP SERVER 1.3.28 Confirm that IBM HTTP Server started successfully by opening a browser and typing in your server name in the address box. Issue the following command from a console window to stop the IBM HTTP Server: NET STOP IBM HTTP SERVER 1.3.28 If the IBM HTTP Server does not start, or you did not accept the default WebSphere Voice Server configuration to install IBM HTTP Server Apache as a Windows service, do the following: 1. Change directory to IBM HTTP Server\bin. 2. Run the apache.exe file from a command line.
143
144
Chapter 7. Tuning
Use the topic below to adjust the configuration and improve the performance of WebSphere Voice Server.
Tuning your configuration

Each installation will have a unique configuration. Tuning and capacity requirements are necessary to support each configuration. For complete information on tuning and WebSphere Application Server IBM HTTP Server, see the Tuning section of the latest WebSphere Application Server 5.1.x Information Center. The parameter values in the following topics were used successfully during product testing when running a large number of sessions. Depending upon your installation configuration, these values might need updating to match your system configuration performance objectives. Use the following only as a guide for possible tuning values.
Tuning WebSphere Voice Server

The WebSphere Voice Server cache default expiration time is 24 hours. This value is used by the HTTP caching policy when Web resources, such as grammar files, do not provide an expiration setting in the HTTP headers, or equivalent metadata. The Web server should be configured with the appropriate expiration information for each resource or group of resources so that the caching performance is optimal. If it is not possible to configure the Web server or other specific resources to provide an appropriate expiration value, then you can change the default expiration setting on WebSphere Voice Server. This value is applied to the entire system so that any resource that does not provide an expiration value is assigned the default. For example, to configure WebSphere Voice Server to expire all undefined resources after eight days, set the value to 691200000 (com.ibm.voice.server.rr.fetchexpires=691200000). This large cache timeout value (in milliseconds) prevents the resources with no set expiration time from expiring before eight days. To set this configuration parameter, use the putParam sample on page 128 script. The following is an example of how to set the parameter on a Windows machine:
cd C:\Program Files\WebSphere\VoiceServer\samples putParam.bat com.ibm.voice.server.rr.fetchexpires 691200000
Tuning WebSphere Voice Server for Avaya, Cisco, and WebSphere Voice Response
Tune the WebSphere Voice Server configuration parameters to the following values: v The number of ASR engines should equal the number of channels. v The number of TTS engines should equal half the number of channels.
145
Note: The number of engines listed above are suitable for initial settings. Your individual configuration will depend on your application requirements.
Tuning WebSphere Voice Server for Genesys

Tune the WebSphere Voice Server configuration parameters to the following values: v The number of ASR engines should equal the number of channels. v The number of TTS engines should equal the number of channels. Note: The number of engines listed above are suitable for initial settings. Your individual configuration will depend on your application requirements. Important: If Genesys TTS fetch hint is set to the default of prefetch and the application uses TTS extensively, such as if it has multiple or back-to-back prompts for each turn, you may need to configure the number of TTS engines to equal twice the number of channels. For the optimal use of TTS resources, it is recommended that you disable Genesys TTS prefetch by setting the value of the VoiceXML extension property name, com.genesys.ttsfetchhint, to safe. For example, in your VoiceXML application, set the following:
<property name="com.genesys.ttsfetchhint" value="safe"></property>
v The RTSP Socket Timeout parameter should equal 0 (com.ibm.voice.server.rtspbridge.sockettimeout=0). This value should already be set by the configureGenesys.sh or configureGenesys.bat file. Also, if you are using Genesys Voice Platform (GVP) 7.0.3 with voice over IP (VoIP) only, and your TTS delay to first audio is long, then configure the TTS audio size using the following steps: 1. Open the Genesys Voice Portal Manager (VPM) and the Server Explorer. Note: You can access the Server Explorer by going to http://localhost/VPM, and then clicking Server Explorer. 2. Expand the link on the left and navigate to the tts_mrcp link. 3. Click the tts_mrcp link. On the right you should see a panel titled tts_mrcp. 4. Click Add New Attribute. The panel refreshes with two new fields. 5. In the Name column, enter minaudiodnldsize, with a value of 2048 and click Save. You now have a new attribute on this screen called Minimum Audio Download Size in Kb and it is set to 2048. 6. Restart the Watchdog for the new value to take effect.
Tuning WebSphere Application Server

The information in this section applies to tuning WebSphere Voice Server version 5.1.2 and earlier. If you are running WebSphere Voice Server 5.1.3, the tuning in this section was applied during your installation. In most cases, the Java Virtual Machine (JVM) Maximum Heap Size should be 256 MB. This is the WebSphere Application Server default. If you are running a large number of automatic speech recognition (ASR) and text-to-speech (TTS) engines, then the JVM Maximum Heap Size should be 512 MB. The JVM Initial Heap Size is 128 MB.
146
For more information on how to change the Maximum Heap Size or Initial Heap Size, refer toJava virtual machine settings in the WebSphere Application Server Information Center.
Tuning AIX
The information in this section applies to tuning WebSphere Voice Server version 5.1.3 only. The tuning in this section was applied during your installation of WebSphere Voice Server 5.1.3. The installation added the following commands to the WebSphere Voice Server section of your /etc/profile file:
ulimit -n 4096
Note: If you have modified the default values of ulimit to be greater than 4096, then your default value will remain in the /etc/profile file. The installation may have added the following commands to the WebSphere Voice Server section of your /etc/rc.net file.
no no no no no -o -o -o -o -o tcp_timewait=2 tcp_nodelayack=1 sb_max=1048576 udp_recvspace=1048576 udp_sendspace=1048576
Make note of the following conditions that affect these command settings: v If tcp_timewait is not changed from the default, then it equals 1 (tcp_timewait=1). If the current value for tcp_timewait is greater than 2, then the setting is changed to 2 (tcp_timewait=2). v If sb_max is less than 1048576, it will be set to in order to 1048576 (sb_max=1048576). v The values for udp_recvspace and udp_sendspace are always set to 1048576 (udp_recvspace=1048576 and udp_sendspace=1048576). In addition, the installation set the following settings for the user root account on which WebSphere Application Server runs:
Soft Soft Soft Soft Hard Hard Hard Hard FILE Size CPU Time STACK Size CORE File Size FILE Size CPU Time STACK Size CORE File Size -1 -1 -1 -1 -1 -1 -1 -1
Important: If you are using an Interactive Voice Response (IVR) system on AIX with WebSphere Voice Server, such as WebSphere Voice Response, make sure that you set no -o tcp_nodelayack=1 in the /etc/rc.net file on the IVR for the best results.
Tuning Linux: Red Hat 3.0 and SUSE 8.0

The information in this section applies to tuning WebSphere Voice Server version 5.1.2 and earlier. If you are running WebSphere Voice Server 5.1.3, the tuning in this section was applied during your installation.
Chapter 7. Tuning
147
Important: The tuning recommended in this section is not persistent for all command windows or after a reboot unless permanently saved or added to the startup profile. Edit your startup file to ensure the commands run on boot up. The following services should be stopped and disabled.
Table 17. Linux services Service description Redundant Array of Independent Disks (RAID) Support Common UNIX Printer System Background Scheduler Send Mail Red Hat Network Support (this service does not apply to SUSE) Service name mdmpd cups crond sendmail rhnsd
To disable these services on Red Hat, use the following steps: 1. Select System Settings Server Settings Services from the main menu to get a list of services. 2. Stop each service, as applicable. 3. Remove the check mark beside each service you want to disable. 4. Save your updated configuration before exiting. To disable these services on SUSE, issue the following command for each service you want to disable, where service_name is the name of a service listed in the table above:
service service_name stop
Note: Disable sendmail on SUSE by issuing: chkconfig sendmail off Remember: The Linux firewall must be disabled for the Interactive Voice Response (IVR) to communicate with WebSphere Voice Server. Set the following minimal values for Linux by issuing the following commands: v ulimit -n 4096 v sysctl -w net.core.rmem_default=8388608 v sysctl -w net.core.wmem_default=8388608 v sysctl -w net.core.rmem_max=16000000 v sysctl -w net.core.wmem_max=10485760 v sysctl -w net.ipv4.tcp_keepalive_time=30 Restriction: Do not set this last tuning parameter when using the Genesys Voice Platform (GVP) NE 6.5.4 or EE 6.5.5.
Tuning threading on SUSE 8.0 with SP2a

Apply the tuning described in this topic only if you are using SUSE 8.0 with service pack 2a (SP2a).
148
The pthread lib in SUSE 8.0 with SP2a limits a process to only 1024 threads. To overcome this limitation, SUSE provides an updated pthread.lib in the UnitedLinux Extension Pack for Lotus Domino. For optimal tuning with SUSE and WebSphere Voice Server, complete the following steps: 1. Download and install the UnitedLinux Extension Pack for Lotus Domino. If the link has moved, perform a search using, UnitedLinux Extension Pack for Lotus Domino for the updated link. 2. Update the wvs.setenv script file provided by the WebSphere Voice Server installation to include the updated pthread.lib. For example:
export LD_LIBRARY_PATH= /opt/dominowrap/domino6:$SPCH_ROOT/bin:$LD_LIBRARY_PATH export LD_PRELOAD= /opt/dominowrap/domino6/libpthread.so:/opt/dominowrap/domino6/librt.so
Tuning Windows Server 2003 Enterprise and Standard editions

The information in this section applies to tuning WebSphere Voice Server version 5.1.2 and earlier. If you are running WebSphere Voice Server 5.1.3, the tuning in this section was applied during your installation. WebSphere Voice Server handles heavy call volumes more efficiently if the following TCP/IP parameters are set in the Windows registry: 1. Start the registry editor: a. Click the Windows Start button. b. Click Run. c. Type regedit in field provided. d. Click OK 2. Use the following directory path to navigate to the registry key:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters
3. In the right pane of the registry editor, look for the TcpTimedWaitDelay value name. If it is not there, add it by selecting Edit New DWORD Value from the menu bar. Type the value name TcpTimedWaitDelay in the name box that appears with the flashing cursor. Note: If you do not see the a flashing cursor and New Value # inside the box, right-click inside the right panel and select Rename from the menu, then type the value name TcpTimedWaitDelay in the name box. 4. Repeat this step for the following value names: MaxUserPort, MaxHashTableSize, and MaxFreeTcbs 5. Double-click inside the right pane again to set the value of TcpTimedWaitDelay. Select Decimal as the Base, and enter 30 in the Value data field. 6. Repeat step 5 for the following value names with the indicated value data: v MaxUserPort = 65534 (Decimal) v MaxHashTableSize = 65536 (Decimal) v MaxFreeTcbs = 16000 (Decimal) See the Microsoft Web site for additional information about setting TCP/IP parameters in the registry.
Chapter 7. Tuning
149
Tuning the Cisco gateway

There are some key Cisco router configurations required to set up and optimize the Cisco Media Resource Control Protocol (MRCP) client. However, in order to configure the Cisco router to recognize a voice server, you must define the MRCP server hosts by issuing the ivr asr-server and ivr tts-server commands. For each command specify an RTSP URL that refers to the host and/or port of the speech server that contains automatic speech recognition (ASR) and/or text-to-speech (TTS) resources. For example, if your WebSphere Voice Server has an IP address of 34.34.130.1, you would issue the following: v ivr asr-server rtsp://34.34.130.1/media/recognizer v ivr tts-server rtsp://34.34.130.1/media/synthesizer
Configuring the HTTP client

The Cisco MRCP client uses an HTTP client, which is a part of the Cisco IOS software. The client fetches VoiceXML documents, audio files, and other file resources. To optimize the fetching of the HTTP client, set the cache configurations. For example, if a 1 MB audio file is being fetched, ensure the audio file is also cached. If there is not enough space configured on the router to hold it, the Cisco HTTP client will fetch the same audio file each time it is needed, causing delays in processing. There are two key HTTP client cache configurations: v http client cache memory file size in kb 1 10,000 Note: The upper range is the maximum file size allowed for caching. Any file over 10 MB will never be cached. v http client cache memory pool size in kb 1 10,000 Note: The upper range is the maximum space allowed for all cached files. The space available can not exceed 10 MB, otherwise subsequent files will not be cached. If you have further questions, or need additional HTTP client configurations, issue the http client ? command for a listing of all available HTTP client configuration commands.
Configuring the MRCP client

The MRCP client has two timeout values that should be configured based on the expected load of your WebSphere Voice Server and the voice application being processed. To avoid premature connection loss on an initial connection, set the following configuration: mrcp client timeout connect time in sec 1 - 20 Setting this value to 5 seconds, should ensure that a connection is made each time. If a connection is not made after 5 seconds, consult the troubleshooting section of the information center for possible resolutions. If the MRCP client encounters an unexpectedly slow response, it may prematurely disconnect during a call. To avoid this situation, set the following configuration: mrcp client timeout message time in sec 1 - 20
150
Note: Setting this value between 6 to 10 seconds should ensure that the WebSphere Voice Server is responding. Any time set over 10 seconds may result in an MRCP client disconnect. If you have further questions, or need additional MRCP client configurations, issue the mrcp client ? command for a listing of all available MRCP client configuration commands.
Configuring the voice port

When configuring a voice port on the Cisco router, set the following configurations so that the router works properly with WebSphere Voice Server: v no echo-cancel enable This value should be set to avoid no-match situations caused by echo cancellation. If this parameter is not set correctly, the WebSphere Voice Server ASR engine receives the echo (which may be prerecorded audio or TTS) and generates a no-match result. In order to get better recognition results, disable echo-cancel. v no vad This value should be set so that the Cisco router voice activity detection is disabled. WebSphere Voice Server already has silence and speech detection. If the Cisco voice activity detection is enabled, then clipped audio is sent to the ASR engine.
Chapter 7. Tuning
151
152
Chapter 8. Troubleshooting
As an activity, troubleshooting or problem determination encompasses a wide range of tasks that might need to be performed at any phase in product usage. In addition to built-in preventative measures, the product provides a variety of tools to make problem determination easier. The troubleshooting section of the documentation helps you understand why your voice server or product installation is not working, and helps you resolve the problem. To effectively troubleshoot a WebSphere Voice Server problem, use the following steps: 1. Understand messages and alarms 2. Understand log messages 3. Enable and monitor trace 4. Collect information for IBM Support You may also troubleshoot according to the component of your WebSphere Voice Server and its environment, or according the particular task you are attempting (such as installation or administration). Additional information on both approaches is also provided to streamline your troubleshooting process: v Troubleshoot by component v Troubleshoot by task type
More troubleshooting resources

v IBM Support Assistant v WebSphere Voice Server support site
Messages and alarms

The purpose of this section is to aid you in understanding WebSphere Voice Server messages. Troubleshooting a deployed system involves collecting and analyzing different types of information from WebSphere Voice Servers many components. The types of information that are gathered and reported are as follows: v Informational messages Action is not required. Communication in this category is for advisory purposes. v Warning messages Recoverable error or event. You should review message, heightened application monitoring recommended, and (as necessary) take any action warranted to avoid future system malfunction. v Error messages Non-recoverable error or event. Action is required to correct problem. Full system function is non operational until the problem is corrected. v Alarms Messages that, besides being logged, are also reported as a console messages to system operator.
153
Several tools provided with WebSphere Voice Server provide messages and alarms to alert administrators about system events: v Administrative console Manage and view log and trace files and view alarms (console messages). The Voice Servers panel automatically refreshes every 60 seconds, but may be configured to refresh on a different schedule. v Voice Trace Analyzer Examine recognition data in multiple trace.log files to build a comprehensive overview of your systems recognition performance and improve the awareness of recognition problems. v WebSphere Application Server system log Errors, warnings, and informational messages. The log file is located at WAS_ROOT/logs/server1/ SystemOut.log. System logs cannot be disabled by a system administrator; they are always created and used anytime the system is operational. Trace logs can be enabled or disabled as needed by a system administrator or by technical support.
Monitoring log messages

WebSphere Voice Server can write system messages to several general purpose logs. WebSphere Voice Server logs its messages to the WebSphere Application Server JVM log (SystemOut.log) and its trace messages to the WebSphere Application Server Trace (trace.log). Both files are located in WAS_ROOT/logs/server1/. Comprehensive information about working with message logs may be found in the WebSphere Application Server Network Deployment 5.1.x Information Center.
Messages by component
This section describes the WebSphere Voice Server messages, which use a 10-character format. CWVSSNNNNT Where: CWV = IBM-registered three-character prefix identifying WebSphere Voice Server. The elements of the CWV code appear below: v SS = Used to identify the WebSphere Voice Server components within the product. See the chart below for more information. v NNNN = Numeric message ID (4-digits). v T = Severity type (Information, Warning, or Error). Click a component identifier below to see all the messages for that component. Note: The message number is any 4-digit number associated with the log message.
Table 18. Component log messages Component identifier CWVAB on page 155 CWVAC on page 160 CWVAE on page 164 Component ASR Bean ASR Control Interface ASR Engine
154
Table 18. Component log messages (continued) Component identifier CWVCC on page 240 CWVGC on page 241 CWVGM on page 244 CWVMC on page 244 CWVMD on page 250 CWVPT on page 252 CWVRB on page 259 CWVRR 0000 - 4999 on page 269 CWVRR 5000 - 9999 on page 270 CWVRT on page 272 CWVSB on page 276 CWVSM on page 281 CWVSY 1000 - 2999, 9000 - 9999 on page 281 CWVSY 3000 - 3999 on page 287 CWVTE on page 289 TTS Engine RTSP Bridge Synthesizer Session Manager System Management and Configuration Component Common Grammar Compiler Grammar Manager Media Converter MRCP Message Dispatcher Product Tools Recognizer Resource Resolver
CWVAB
CWVAB0001E: ASR allocation failed: VALUE_0
Explanation: Failed to allocate an ASR resource. User response: Check for additional errors in the SystemOut.log file to further diagnose the problem. Refer to Troubleshooting section of the Information Center for possible solutions.
CWVAB0002E: Error parsing ASR recognition results: VALUE_0

Explanation: An error occurred while parsing recognition results. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0003E: ASR getParam failed: VALUE_0

Explanation: An error occurred while processing a request to retrieve a parameter from the ASR resource. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/
155
index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0004E: ASR setParam failed: VALUE_0

Explanation: An error occurred while processing a request to change a parameter in the ASR resource. A communication error occurred between the ASR Bean and ASR Control components. Review the error message for details. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0005E: ASR getParams failed: VALUE_0

Explanation: An error occurred while processing a request to retrieve parameters from the ASR resource. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0006E: ASR setParams failed: VALUE_0

Explanation: An error occurred while processing a request to change parameters in the ASR resource. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0007E: ASR loadGrammars failed: VALUE_0

Explanation: An error occurred while attempting to load recognition grammars. Insure system memory resources are sufficient for loading the grammar in the Administration Console (Status Panel). User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
156
CWVAB0008E: ASR recognize request failed: VALUE_0

Explanation: An error occurred while attempting to start a speech recognition session. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0009E: ASR startRecognitionTimer request failed: VALUE_0

Explanation: An error occurred while requesting the ASR resource to start the no-input timers. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0011E: ASR Control message syntax error: VALUE_0

Explanation: This is an unexpected error. The cause cannot be immediately determined. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0012E: ASR Control message I/O error: VALUE_0

CWVAB0014E: I/O error - disconnecting VALUE_0

Explanation: This is an unexpected error. The cause cannot be immediately determined. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is
157
available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0015E: Internal Request failed: VALUE_0, reason=VALUE_1, beanInfo=VALUE_2

Explanation: A communication failure occurred. However, the cause cannot be immediately determined. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAB0016E: Exception reading internal configuration, using default values: VALUE_0

Explanation: An error occurred while attempting to read WVS internal configuration. User response: Refer to the Troubleshooting section of Information Center to verify the installation completed successfully.
CWVAB0017E: Invalid ASR Engine codepage VALUE_0 - using default character set VALUE_1
CWVAB0018W: ASR Engine codepage not provided, using default character set VALUE_0
158
CWVAB0019E: Charset is not supported in the Java Virtual Machine: [VALUE_0]

CWVAB0020E: Configuration error: Property VALUE_0 is not set

CWVAB0021E: Configuration error: Invalid configuration: VALUE_0=VALUE_1

CWVAB0022E: Encoder error: VALUE_0

CWVAB0023E: Internal Request failed: VALUE_0, beanInfo=VALUE_1

159
CWVAB0026E: Stop request failed: VALUE_0

CWVAB0027E: Grammar unloading failure: VALUE_3. Grammar:VALUE_0,VALUE_1,VALUE_2

Explanation: An error occurred while unloading a grammar in the ASR engine. This error can cause grammar memory resources to remain allocated by the system. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAC
CWVAC0015E: VALUE_0 Socket error detected on a VALUE_1 function call - return code: VALUE_2, reason: VALUE_3
CWVAC0016E: VALUE_0 Errors detected during ASRControl processing

160
CWVAC0018E: VALUE_0 Unable to allocate VALUE_1 resource

CWVAC0019W: VALUE_0 ASR Request message is missing VALUE_1 parameter

CWVAC0020E: VALUE_0 Caught an unhandled exception. ASRControl process exiting

CWVAC0021E: VALUE_0 VALUE_1 allocation failure

CWVAC0026E: VALUE_0 VALUE_1 component connection failed at host:VALUE_2, port:VALUE_3

Explanation: This is an unexpected error. The cause cannot be immediately determined.
161
User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAC0027E: VALUE_0 Invalid value VALUE_1 for VALUE_2 parameter

CWVAC0028E: VALUE_0 Missing connection information to VALUE_1

CWVAC0029E: VALUE_0 Invalid connection VALUE_1 information ==> VALUE_2

CWVAC0030E: VALUE_0 Invalid VALUE_1 format

162
CWVAC0031E: VALUE_0 Invalid ASRControl Message Handler State VALUE_1 for VALUE_2 request
CWVAC0032E: VALUE_0 Invalid Request received from ASRBean: VALUE_1

CWVAC0033E: VALUE_0 Unrecoverable condition detected during ASR processing - ASRControl process preparing for shutdown
CWVAC0036E: VALUE_0 VALUE_1 data is missing from the body of the ASR Request message
163
CWVAE
CWVAE0001W: %s %s: Failed to allocate storage for %s
Explanation: Insufficient virtual memory for the process. User response: Check available paging space.
CWVAE0002W: %s %s: Received invalid parameter for start or end time

CWVAE0003W: %s %s: Received invalid parameter for %s

CWVAE0004W: %s %s: Failed to open file %s

CWVAE0005W: %s %s: Received error return code %d from %s

164
CWVAE0006W: %s %s: Haifa compression is not supported

CWVAE0007W: %s %s: Migrating old pool data

CWVAE0008W: %s %s: Bad value %s for tag %s in file %s

CWVAE0009W: %s %s: Bad data in file %s

CWVAE0010W: %s %s: Exceeded limit of %d pools

165
CWVAE0011W: %s %s: Exceeded limit of %d baseforms in pool

CWVAE0012W: %s %s: Deleting pool set %d in use

CWVAE0013W: %s %s: Deleting pool %d in use

CWVAE0014W: %s %s: Attempting to release missing pool %d

CWVAE0015W: %s %s: No baseform translator available

166
CWVAE0016W: %s %s: Invalid phone type %s in file %s

CWVAE0017W: %s %s: Number of baseform spellings %d does not match number of offsets %d
CWVAE0018W: %s %s: Failed to load sounds like spelling for %s, number of baseforms do not match
CWVAE0019W: %s %s: Failed to write file %s

Explanation: Failed to write the file. User response: Check for file system full or improper permissions on the directory or filesystem.
CWVAE0020W: %s %s: Baseform for %s has an illegal phone 0X%x

167
CWVAE0021W: %s %s: Failed to save baseform for %s, errno %d

CWVAE0022W: %s %s: Invalid number of phones in file %s

CWVAE0023W: %s %s: Exceeded limit of %d liasons

CWVAE0024W: %s %s: Invalid phone nunber for liason phone %s

168
CWVAE0025W: %s %s: Invalid phone nunber for acceptor phone %s

CWVAE0026W: %s %s: Questions file %s length %d is not a multiple of the number of phones %d.
CWVAE0027W: %s %s: Questions file %s length %d is not a product of the number of phones %d and the number of questions %d
CWVAE0028W: %s %s: Number of trees %d in %s does not match number in header %d from %s
CWVAE0029W: %s %s: A tree from %s exceeds the number of arcs %d

169
CWVAE0030W: %s %s: Out-of-range arc number in %s

CWVAE0031W: %s Caught signal %s (%d), program %s, routine %s

CWVAE0032E: %s ASR Engine terminating on signal %s (%d)

CWVAE0033W: %s %s: Attempting to remove missing pool %d

170
CWVAE0035W: %s %s: Exceeded limit of %d clients

CWVAE0036W: %s %s: Attempt to add existing client 0x%x rejected

CWVAE0037W: %s %s: Invalid phase set %d

CWVAE0039W: %s %s: Illegal baseform %d for word %s in vocabulary %s

CWVAE0041W: %s %s: failed to insert non-acoustic sel for word %d in vocabulary %d

171
CWVAE0042W: %s %s: Mode specified is 0x%x, but no personal or temp pool exists
CWVAE0043W: %s %s: Failed to add spelling %s to pool

CWVAE0044W: %s %s: Acoustic Addword not enabled, no Acoustic-File entry in task file
CWVAE0045W: %s %s: Acoustic Addword only using the first of the %d utterances specified
Explanation: This is an unexpected error. The cause cannot be immediately determined. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/
172
CWVAE0046W: %s %s: No temporary pool available for acoustic addword of spelling %s

CWVAE0047W: %s %s: Failed to add spelling %s to temp pool for acoustic addword
CWVAE0048W: %s %s: Engine not halted when adding spelling %s

CWVAE0049I: %s ASR Engine 0x%x is open for on demand e-business

Explanation: Normal operation. There is one message for every speech recognition engine started. User response: No action required.
CWVAE0050I: %s ASR Engine 0x%x is shutting down

Explanation: Normal operation. There is one message for every speech recognition engine shut down. User response: No action required.
173
CWVAE0051W: %s %s: Failed to auto-add spelling %s, return code %d

CWVAE0055W: %s %s: Failed to save baseform for spelling %s

CWVAE0056W: %s %s: Failed to auto-addword using sounds-like spelling %s, return code %d
CWVAE0057W: %s %s: Failed to make word from file %s, return code %d
CWVAE0058W: %s %s: Engine not halted when processing spelling %s

174
CWVAE0059W: %s %s: Invalid tags specified when processing spelling %s

CWVAE0060W: %s %s: Discarded tag %d specified when processing spelling %s

CWVAE0061W: %s %s: Specified tags span utterance boundary when processing spelling %s
CWVAE0062W: %s %s: Specified tag (%d:%d) not associated with any vocabulary owned by client when processing spelling %s
175
CWVAE0063W: %s %s: Vocabulary %s associated with tag (%d:%d) has no appendix when processing spelling %s
CWVAE0064W: %s %s: Specified tags (%d:%d) not found in tag file

CWVAE0065W: %s %s: Invalid null string specified

CWVAE0066W: %s %s: Invalid character in sounds-like string %s

176
CWVAE0067W: %s %s: Failed to update spelling %s in vocabulary %s

CWVAE0068W: %s %s: Decoded has not been initialized

CWVAE0069W: %s %s: Request to iterate made with mike off

CWVAE0070W: %s %s: Error return code %d from dc_iter, forcing microphone off
CWVAE0071W: %s %s: Failed to define vocabulary for silence word %s. Baseform may be missing or empty
177
CWVAE0072W: %s %s: The vocabulary name %s is already in use

CWVAE0073W: %s %s: Failed to load prototypes for acoustic identifier %s

CWVAE0074W: %s %s: Failed to add pool %s, return code %d

CWVAE0075W: %s %s: Failed to add vocabulary %s

178
CWVAE0076W: %s %s: Failed to define static vocabulary %s

CWVAE0077W: %s %s: Failed to define appendix vocabulary %s

CWVAE0078W: %s %s: Failed to define a static vocabulary specified in file %s

CWVAE0079W: %s %s: Failed to initialize label buffer

CWVAE0080W: %s %s: Failed to add words to vocabulary %s, return code %d

179
CWVAE0081W: %s %s: Failed to remove words from vocabulary %s, return code %d
CWVAE0082W: %s %s: Exceeded rank limit of %d

CWVAE0083W: %s %s: Stanza name at index %d in file %s is null

CWVAE0084W: %s %s: Tag name at index %d of stanza %s in file %s is null

180
CWVAE0085W: %s %s: Failed to delete file %s

CWVAE0086W: %s %s: No language specified in enrollid %s for userid %s

CWVAE0087W: %s %s: Specified userid %s does not exist

CWVAE0088W: %s %s: Failed to read file %s

CWVAE0089W: %s %s: Failed to define %d lexemes for vocabulary %s

181
CWVAE0090W: %s %s: Invalid phone %d at index %d in baseform %d from pool %s

CWVAE0091W: %s %s: End of utterance, ignoring setalarm at time %d

CWVAE0092W: %s %s: Request to read labels already processed at time %d

CWVAE0093W: %s %s: Cache file %s has invalid format

182
CWVAE0094W: %s %s: Invalid vocabulary number %d specified

CWVAE0095W: %s %s: No vocabularies exist

CWVAE0096W: %s %s: Invalid language model type %d specified

CWVAE0097W: %s %s: Failed to rename file %s

CWVAE0098W: %s %s: No static vocabularies specified in task file %s

Explanation: This is an unexpected error. The cause cannot be immediately determined. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the
183
problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVAE0099W: %s %s: Required tag %s is missing in file %s

CWVAE0100W: %s %s: Failed to find langs directory in path %s

CWVAE0101W: %s %s: Invalid language model version %d specified in file %s

CWVAE0102W: %s %s: Language model boundary word %d is not in active vocabulary %s

184
CWVAE0103W: %s %s: Invalid likepath %d in path %d, vocabulary %d

CWVAE0104W: %s %s: Failed to accept socket connection on listen socket %d, errno %d
CWVAE0105W: %s %s: Failed to read socket %d of connecting client, errno %d

CWVAE0106W: %s %s: Received unexpected message type %d from connecting client

CWVAE0107W: %s %s: Failed to read socket %d of connected client 0x%x, errno %d

185
CWVAE0108E: %s %s: Failed to read socket %d of internal connection, errno %d

CWVAE0109E: %s %s: Failed to establish addressability to engines mailbox

CWVAE0110E: %s %s: Failed to load audio library %s

CWVAE0111E: %s %s: Failed to resolve critical function address from library %s

186
CWVAE0112E: %s %s: Failed to load audio/video library %s

CWVAE0113E: %s %s: Failed to initialize audio source, return code %d

CWVAE0114W: %s %s: Missing audio video library name, processing audio only
CWVAE0115W: %s %s: Audio visual dimension %d does not match feature dimension %d
187
CWVAE0116W: %s %s: V dimensionality %d does not match MSHMM visual prototype dimension %d
CWVAE0117E: %s %s: Invalid sample rate %4.1f returned by audio library

CWVAE0118W: %s %s: Initialization of audio compression failed, return code %d - continuing without audio compression
CWVAE0119E: %s %s: Failed to allocate storage for %s

CWVAE0120E: %s %s: Invalid label source %d specified

188
CWVAE0121E: %s %s: Failed to begin recording, audio library return code %d

CWVAE0125W: %s %s: Audio buffer overflow, audio buffer size %d, audio bytes returned %d
CWVAE0126W: %s %s: All label buffers full, labels written %d

CWVAE0127W: %s %s: No audio input file specified

189
CWVAE0128W: %s %s: Failed to begin playback - audio library return code %d

CWVAE0129W: %s %s: Failed to seek file %s

CWVAE0130W: %s %s: Invalid signature %s in file %s

CWVAE0131W: %s %s: Initialization of audio decompression failed, return code %d

CWVAE0132W: %s %s: Exceeded limit of %d characters in a filename

190
CWVAE0133W: %s %s: Invalid data format in package file %s

CWVAE0134W: %s %s: Mapped file %s already exists

CWVAE0135W: %s %s: Invalid index %s in package file %s

CWVAE0136W: %s %s: Value %g of tag %s in file %s is out of range

191
CWVAE0138W: %s %s: Specified language %s is not installed

CWVAE0139W: %s %s: Invalid spellcheck version %d specified in file %s

CWVAE0140W: %s %s: Failed to copy file %s

CWVAE0141W: %s %s: Failed to change access to file %s

CWVAE0142W: %s %s: Number of bands %d is greater than dimension %d

192
CWVAE0143W: %s %s: Failed to create directory %s

Explanation: Failed to create the directory. User response: Check the file system permissions and available space.
CWVAE0144W: %s %s: Invalid last character in spellings buffer

CWVAE0145W: %s %s: Map buffer contains %d bytes but expected %d (4 * number of words)
CWVAE0146W: %s %s: Specifying a map file is valid only when creating new files
CWVAE0147W: %s %s: Failed unordered add without a map of %s to %s

193
CWVAE0148W: %s %s: Invalid request to add %s to read-only word list %s

CWVAE0149W: %s %s: Word list %s must be sorted

CWVAE0150W: %s %s: Bad spellings, offsets, or map file

CWVAE0151W: %s %s: Cant change Select verb via re-define

194
CWVAE0152W: %s %s: Cant change select grammar options via re-define

CWVAE0153W: %s %s: Failed to find CDCN initialization file %susing defaults

CWVAE0154W: %s %s: Failed to build c0 histogram with %d samples limited (%f to %f)
CWVAE0155W: %s %s: Error, %s, mapping file %s

CWVAE0156W: %s %s: FM define failed with return code %d for %s

195
CWVAE0157W: %s %s: FM update failed with return code %d for %s

CWVAE0158W: %s %s: FSG data size is not as expected in file %s

CWVAE0159W: %s %s: Failed to build the FM tree

CWVAE0161W: %s %s: Failed to write %s vector to output file %s

196
CWVAE0162W: %s %s: Format 2 or 3 file %s has wrong byte order

CWVAE0163W: %s %s: Found %d words when re-writing %s expected %d

CWVAE0164W: %s %s: Grammar %s has no words

CWVAE0165W: %s %s: Grammar %s has non-acoustic baseform(s)

CWVAE0166W: %s %s: Invalid FSG file %s

197
CWVAE0167W: %s %s: Invalid grammar options %x

CWVAE0168W: %s %s: Invalid select grammar options %x

CWVAE0169W: %s %s: LM define failed with return code %d for %s

CWVAE0170W: %s %s: Missing external %s needed by vocab %s

198
CWVAE0171W: %s %s: Missing pronunciations for external vocabulary %s in file %s needed by vocabulary %s
CWVAE0172W: %s %s: More than %d baseforms in a pool

CWVAE0173W: %s %s: More than %d words in a vocabulary

CWVAE0174W: %s %s: Output dimension %d should be greater than or equal to %d

CWVAE0175W: %s %s: Output dimension %d should be less than or equal to %d

199
CWVAE0176W: %s %s: Problem reading FSG header in file %s

CWVAE0177W: %s %s: Return code %d trying to build the FM tree

CWVAE0178W: %s %s: Select vocab needs more than just verbs

CWVAE0179W: %s %s: Unknown word number %d will be ignored

200
CWVAE0180W: %s %s: Vocabulary V%d s not sorted

CWVAE0181W: %s %s: Vocabulary %s is already defined, but needs to be defined as an external by vocabulary %s
CWVAE0182W: %s %s: Will leak memory! Resolve should be managing buffers

CWVAE0183W: %s %s: Wrong sized initial %s vector in file %s, %d != %d

CWVAE0184E: %s %s: FFT data allocation is not 16-byte aligned for %s

201
CWVAE0185W: %s %s: More than %d vocabularies defined

CWVAE0186W: %s %s: Failed to access vocabulary %d

CWVAE0187W: %s %s: Invalid destination vocabulary number %d

CWVAE0188W: %s %s: Invalid destination vocabulary %d

202
CWVAE0189W: %s %s: Have enabled a different do-care vocabulary V%d

CWVAE0190W: %s %s: More than %d prototype sets defined

CWVAE0191W: %s %s: Failed with return code %d reading %s file

CWVAE0192W: %s %s: Failed with return code %d reading file %s

CWVAE0193W: %s %s: Failed to resolve %s

203
CWVAE0194W: %s %s: Failed with return code %d opening file %s

CWVAE0195W: %s %s: Found %d columns in file %s but expected %d

CWVAE0196W: %s %s: Found %d rows in file %s but expected %d

CWVAE0197W: %s %s: Invalid number of columns read (%d) from file %s

204
CWVAE0198W: %s %s: Read %d bytes from %s, but expected %d, return code %d
CWVAE0199W: %s %s: Number of rows (%d) in file %s is less than the minimum rank position (%d or %d)
CWVAE0200W: %s %s: Expected %d clusters in map, but found %d in %s

CWVAE0201W: %s %s: Cluster selection, %d, exceeds maximum of %d

CWVAE0202W: %s %s: Failed to map selection prototypes (%s) from file %s

205
CWVAE0203W: %s %s: Failed to map shared acoustic image (%s) from file %s
CWVAE0204W: %s %s: Failed to read statistics for file %s

CWVAE0205W: %s %s: Mismatched prototypes %s%s

CWVAE0206W: %s %s: Version number, %s, exceeds maximum of 99 for file %s

206
CWVAE0207W: %s %s: Acoustic ID name %s is too long

CWVAE0208W: %s %s: Buffer overrun. Lost %d mel feature vectors

CWVAE0209W: %s %s: Failed to determine the name of the initial mean normalization file
CWVAE0210W: %s %s: Failed to find directory for userid %s

CWVAE0211W: %s %s: Failed to initialize FFT of size %d

207
CWVAE0212W: %s %s: Failed to initialize HRLPitchDetector

CWVAE0213W: %s %s: Failed to load audio-visual clustered prototypes

CWVAE0214W: %s %s: Failed to migrate EID file %s

CWVAE0215W: %s %s: Failed to reload acoustic ID %s

208
CWVAE0216W: %s %s: Mean vector has length %d, but it should be %d

CWVAE0217W: %s %s: Unsupported audio format (%d) specified

CWVAE0218W: %s %s: Failed to initialize %s[%d]

CWVAE0219W: %s %s: Failed to initialize prototype mapping

CWVAE0220W: %s %s: Invalid PCA matrix

209
CWVAE0221W: %s %s: Invalid data reading mean vectors

CWVAE0222W: %s %s: Invalid data reading phone to class map

CWVAE0223W: %s %s: Invalid data reading priors

CWVAE0224W: %s %s: Invalid data reading prototype to phone map

210
CWVAE0225W: %s %s: Invalid data reading variances

CWVAE0226W: %s %s: Problem reading PCA matrix in file %s

CWVAE0227E: %s %s: The question is not in the range 0-15

CWVAE0228W: %s %s: Attempted to use disabled function to build baseforms

CWVAE0229W: %s %s: Failed detailed match. Return code %d

211
CWVAE0230W: %s %s: Failed fast match. Return code %d

CWVAE0232W: %s %s: Failed to initialize baseform generation. Return code %d

CWVAE0233W: %s %s: Invalid fast match tree. Return code %d

CWVAE0234W: %s %s: Invalid addword question found at offset %d

212
CWVAE0235W: %s %s: Soundslike spelling has only NULL phone letters

CWVAE0236W: %s %s: The next letter sequence is not in the range 0-31
CWVAE0237W: %s %s: Non-acoustic sel failed to insert

CWVAE0238W: %s %s: Given sel %d# with illegal vocabulary V%d

CWVAE0239W: %s %s: Attempting to delete overridden sel %d (word %d V%d) with %d extensions
213
CWVAE0240W: %s %s: Bad data in %s

CWVAE0241W: %s %s: Failed to apply rules to phone string %s

CWVAE0242W: %s %s: Failed to find .fst file in path %s

CWVAE0243W: %s %s: Failed to find Resolve file

214
CWVAE0244W: %s %s: Failed to initialize Letter Output List - return code %d

CWVAE0245W: %s %s: Failed to initialize Letter Outputs. Return code %d

CWVAE0246W: %s %s: Failed to initialize phones file %s

CWVAE0247W: %s %s: Failed to load FST grammar %s

CWVAE0248W: %s %s: Failed to load grammar library %s

215
CWVAE0249W: %s %s: Failed to load sounds-like library %s

CWVAE0250W: %s %s: Failed with return code %d initializing data files

CWVAE0251W: %s %s: Failed to open handle to client %s process

CWVAE0252W: %s %s: Failed to add client, exceeded maximum of %d clients

216
CWVAE0253W: %s %s: Failed to remove client, specified handle %d is not valid

CWVAE0254W: %s %s: Specified enrollid %s does not exist for userid %s

CWVAE0255W: %s %s: Specified taskid %s does not exist

CWVAE0256W: %s %s: Specified topic %s does not exist

CWVAE0257W: %s %s: Specified password is not valid

217
CWVAE0258W: %s %s: Specified taskid %s does not exist in language %s

CWVAE0259W: %s %s: Taskid language %s does not match enrollid language %s

CWVAE0260W: %s %s: Specified userid %s does not match currently initialized userid %s
CWVAE0261W: %s %s: Specified enrollid %s does not match currently initialized enrollid %s.
218
CWVAE0262W: %s %s: Specified taskid %s does not match currently initialized taskid %s
CWVAE0263W: %s %s: Specified audio type %s does not match currently initialized audio type %s
CWVAE0264W: %s %s: Specified audio library %s does not match currently initialized audio library %s
CWVAE0265W: %s %s: Failed to list available scripts, return code %d

219
CWVAE0266W: %s %s: No script available which supports unsupervised adaptation

CWVAE0267W: %s %s: Failed to create unsupervised adaptation directory

CWVAE0268W: %s %s: No tag file exists for specified tag 0x%x

CWVAE0269W: %s %s: Failed to update script %s status, return code %d

CWVAE0270W: %s %s: Failed to remove directory %s

220
CWVAE0271W: %s %s: No audio input file found for utterance %d

CWVAE0272W: %s %s: No default enrollid exists for userid %s.

CWVAE0273W: %s %s: No userid has been specified

CWVAE0274W: %s %s: Reserved flags have been set to 0x%x by client 0x%x
221
CWVAE0275W: %s %s: Invalid format in archive file %s, header id %s, header version %d
CWVAE0276W: %s %s: Specified userid %s does not match userid in archive %s

CWVAE0277W: %s %s: Uncompressed archive %s to size %d, but expected size %d

CWVAE0278W: %s %s: Concurrent recognition session not supported with current enrollment session
222
CWVAE0279W: %s %s: Requested userid %s is currently being used exclusively by another application
CWVAE0280W: %s %s: Failed to initialize decoder, return code %d

CWVAE0281W: %s %s: Failed to initialize acoustic processor, return code %d

CWVAE0282W: %s %s: Failed to unlock exclusive user lock in directory %s

Explanation: A lock is being held by a currently running speech recognition engine or by an engine that has crashed. User response: If from an engine that has crashed, remove the lock manually.
CWVAE0283W: %s %s: Tags must be sorted in increasing order

223
CWVAE0284W: %s %s: No spelling has been specified

CWVAE0285W: %s %s: Concurrent enrollment sessions not supported

CWVAE0286W: %s %s: No script has been specified

CWVAE0287W: %s %s: Enrollment session not supported while training is running

CWVAE0288W: %s %s: No enrollid has been specified

224
CWVAE0289W: %s %s: Specified script %s does not exist

CWVAE0290W: %s %s: Specified script %s contains no sentences

CWVAE0291W: %s %s: Failed to create enrollid %s for userid %s

CWVAE0292W: %s %s: Failed to query script %s status

225
CWVAE0293W: %s %s: No task specified in script %s

CWVAE0294W: %s %s: Specified utterance %d is out of range (0 - %d) for script %s

CWVAE0295W: %s %s: Failed to query status of enrollid %s for userid %s

CWVAE0296W: %s %s: Starting training while training is running is not supported

CWVAE0297W: %s %s: Failed to launch training program

226
CWVAE0298E: %s %s: Failed to create listening socket on port %d, errno %d

Explanation: Failed to create the specified TCP/IP listening socket. User response: Verify the specified port is not being used by another program.
CWVAE0299E: %s %s: Failed to connect to listening socket for internal messages, errno %d
CWVAE0300E: %s %s: Failed to accept connection for internal messages, errno %d

CWVAE0301E: %s %s: Current run directory %s is not accessible, errno %d

Explanation: Failed to read the directory in which the speech recognition engine is running. User response: Check the file system permissions.
CWVAE0302W: %s %s: No handler defined for message %d

227
CWVAE0303W: %s %s: Session save or restore operation in progress

CWVAE0304W: %s %s: Client 0x%x is already attached

CWVAE0305W: %s %s: Current api version %s does not match client api version %s
CWVAE0306W: %s %s: Current run directory %s does not match specified run directory %s
228
CWVAE0307W: %s %s: Current services %s do not match requested services %s

CWVAE0308E: %s ASR engine 0x%x is terminating

CWVAE0309W: %s %s: Disk space required for training (%4.1f), exceeds available disk space (%4.1f)
CWVAE0310W: %s %s: Failed to create thread for function %s

CWVAE0311W: %s %s: No vocabulary has been specified

229
CWVAE0312W: %s %s: Failed to write file %s to archive %s

CWVAE0313W: %s %s: Failed to calculate CRC for archive %s

CWVAE0314W: %s %s: Calculated CRC value 0x%x does not match stored CRC value 0x%x in archive %s.
CWVAE0315W: %s %s: Failed to read file from archive %s for utterance %d

230
CWVAE0316W: %s %s: Failed to determine size of file %s

CWVAE0317W: %s %s: Specified utterance number %d is invalid

CWVAE0318W: %s %s: No audio available after 10 seconds - forcing microphone off

Explanation: The ASR engine stopped listening for audio after ten seconds because it did not receive any audio input from the Media Converter. User response: This message indicates a problem with the Interactive Voice Response (IVR) system. Contact support for the IVR system or application you are using.
CWVAE0319W: %s %s: Specified input identifier 0x%x is not valid

CWVAE0320W: %s %s: A dispatch handler is already registered for application %s

231
CWVAE0321W: %s %s: Failed to determine asynchronous notification socket handle for application %s
CWVAE0322W: %s %s: Unexpected return code %d (%d) from select

CWVAE0323W: %s %s: No input sources need to be dispatched

CWVAE0324W: %s %s: Failed to get current time, return code %d.

232
CWVAE0325W: %s %s: A SMAPI session is not open. Unable to send requests or receive replies
CWVAE0326W: %s %s: Failed to dispatch message. Return code %d

CWVAE0327W: %s %s: Unknown message type %d received for application %s

CWVAE0328W: %s %s: There is no connection with the speech recognition engine

CWVAE0329W: %s %s: SM Session API send or receive error: %d, disconnecting from engine
233
CWVAE0330W: %s %s: Ignoring unsupported argument %s

CWVAE0331W: %s %s: Failed to establish a connection with the speech recognition engine, connect code %c
CWVAE0332W: %s %s: Failed to execlp %s, return code %d, errno %d

CWVAE0400W: %s %s: Invalid character %c (0x%x) in sounds-like spelling

234
CWVAE0401W: %s %s: Invalid separation character %c (0x%x)

CWVAE0402W: %s %s: String %s is composed of all separator characters

CWVAE0403W: %s %s: Unable to find phones.

CWVAE0404W: %s %s: Missing start boundary

CWVAE0405W: %s %s: Missing end boundary

235
CWVAE0406W: %s %s: Failed detailed match for silence at (%d %d %d) with return code %d
CWVAE0407W: %s %s: LM-Class-Maps is no longer supported

CWVAE0408W: %s %s: Failed to map file %s (%s)

CWVAE0409W: %s %s: LM n-gram file size is %d but the index file expects %d
236
CWVAE0410W: %s %s: Obsolete LM n-gram file

CWVAE0411W: %s %s: Invalid parameter name %s

CWVAE0412W: %s %s: Invalid value (%s) for %s

CWVAE0413E: %s %s: Failed to allocate %d bytes

CWVAE0414E: %s %s: Failed to allocate storage

CWVAE0415W: %s %s: Failed to create Tphon rules

237
CWVAE0416W: %s %s: Failed to create Onset rules

CWVAE0417W: %s %s: Failed to create Trailer rules

CWVAE0418W: %s %s: Failed to create Glottal rules

CWVAE0419W: %s %s: Missing ] in definition of phone %s

CWVAE0420W: %s %s: Missing > in definition of class %s

238
CWVAE0421W: %s %s: Failed to find class for %s

CWVAE0422W: %s %s: Failed to load topic %s

CWVAE0423W: %s %s: Failed to find phone for %s

CWVAE0424W: %s %s: Illegal baseform %d for word %d in undefined V%d

239
CWVAE0425W: %s %s: Found %d alloarcs for baseform %d, word %s in vocabulary %s

CWVAE0426W: %s %s: Found %d phones for baseform %d

CWVCC
CWVCC0001E: Failed to register control channel with selector channel was closed - port number: VALUE_0
CWVCC0002E: Atypical unrecoverable error in selector - leaving listener thread - reason: VALUE_0
CWVCC0003E: I/O exception reading message - reason: VALUE_0

240
CWVCC0004E: Error occurred when parsing message - reason: VALUE_0

CWVCC0005E: Native Error Stream: VALUE_0

Explanation: Information written by a native process to the error stream. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVCC0006I: OutboundMessageFilter header filename loaded: VALUE_0

Explanation: Informs which message filter has been loaded. User response: No user action required.
CWVGC
CWVGC2001E: Compiler property name is null
Explanation: Error occurred while setting the compilers property. User response: No user action needed.
CWVGC2002E: Cannot set null value for Compiler property VALUE_0

241
CWVGC2003E: VALUE_0 is an illegal value for compiler property VALUE_1

CWVGC2004E: Compilers object property name is null

Explanation: Error occurred while setting the compilers object property. User response: No user action needed.
CWVGC2005E: Cannot set null value for compilers object property VALUE_0
CWVGC2006E: VALUE_0 is an illegal value for compilers object property VALUE_1

CWVGC2021E: empty mediaType argument

Explanation: The grammar document media-type cannot be empty. User response: No user action needed.
CWVGC2031E: Invalid state. No grammar loaded

Explanation: The compiler is called for compiling a grammar before loading the grammar or loading a grammar has failed. User response: No user action needed.
CWVGC2032E: Language not specified for voice grammar

Explanation: Language is required to compile a grammar. User response: No user action needed.
CWVGC2033E: Missing country code on voice grammar language: VALUE_0

Explanation: The language for the grammar should include the country code. User response: No user action needed.
242
CWVGC2034E: Missing language code on voice grammar language: VALUE_0

Explanation: The language for the grammar should include the language and the country code. User response: No user action needed.
CWVGC2035E: VALUE_0 property must be set to build package for voice grammar
Explanation: Cannot build a compiled grammar package without the required property. User response: No user action needed.
CWVGC2036E: Invalid Rule: VALUE_0 in VALUE_1 grammar

Explanation: A rule in the grammar cannot be found. User response: Correct the grammar to include the missing rule.
CWVGC2037E: Unable to build BGF Data for rule: VALUE_0

Explanation: The grammar has logical errors User response: No user action needed.
CWVGC2038E: Compilation/packaging error(s) from native code: VALUE_0

Explanation: Internal error(s) while compiling/packaging the grammar. User response: No user action needed.
CWVGC2039E: System Error in compilation/packaging

Explanation: System error(s) while compiling/packaging the grammar. User response: No user action needed.
CWVGC2040E: Illegal call - cannot compile a pre-compiled grammar

Explanation: The compiler is called for compiling a pre-compiled grammar. User response: No user action needed.
CWVGC2041E: Stack overflow when compiling grammar (rule too long?)

Explanation: Stack overflow when compiling grammar (rule too long?) User response: Try to reduce the length of sequential tokens of the alternatives of the rules.
243
CWVGC4000W: Warning messages from native code: VALUE_0

Explanation: Warning(s) while compiling/packaging the grammar. User response: No user action needed.
CWVGM
CWVGM1000E: Grammar document declares invalid meta data
Explanation: The application provided grammar declares invalid meta data attributes using the meta tag. User response: Ensure the syntax conforms to the specified SRGS grammar format.
CWVGM1001E: Unable to update grammar cache entry

Explanation: The grammar cache entry cannot be updated with the meta data. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVGM1002E: Invalid grammar document URI

Explanation: The format of the application URI is invalid. User response: Ensure the syntax conforms to the URI RFC.
CWVGM1003I: Unable to obtain HTTP header information for resource : VALUE_0

Explanation: The resource is either inaccessible from WVS or the HTTP server does not accept the HEAD request. User response: Specify an accessible URI or an HTTP server that supports HEAD.
CWVMC
CWVMC0001W: Failed to load configuration parameter VALUE_0 defaulting to VALUE_1
244
CWVMC0002E: The Media Converter TTS subsystem was unable to send audio output to the RTP host VALUE_0 and port VALUE_1
CWVMC0003W: The Media Converter TTS subsystem experienced VALUE_0 underrun(s) of audio data from the TTS engine. At time to send TTS output - no audio data was available - additional info follows: media server port=VALUE_1, media client port=VALUE_2
Explanation: The WebSphere Voice Server may be over capacity and not able to produce the RTP audio stream for the SYNTHESIZER resource at a consistent interval to the MRCP client. User response: Verify that the tuning parameters for your operating system are applied as described in the section on tuning operating systems in the WebSphere Voice Server Information Center. Also, reduce the number of expected speech resources or lessen the MRCP client load on WebSphere Voice Server.
CWVMC0004W: The Media Converter TTS subsystem was unable to create audio trace file VALUE_0
CWVMC0005W: The Media Converter TTS subsystem was unable to send a speak complete message to the TTS engine process
Explanation: This warning can occur if an MRCP session teardown is issued to the synthesizer while a turn of audio is ending. User response: If the MRCP synthesizer session was undergoing a teardown when this occurred - ignore the warning.
CWVMC0006W: The Media Converter TTS subsystem was unable to write to audio file trace VALUE_0
245
CWVMC0007W: The Media Converter was unable to obtain the ASR installation config value
CWVMC0008W: The Media Converter was unable to obtain the TTS installation config value
CWVMC0009W: The Media Converter was unable to obtain the ASR buffer size config value
CWVMC0010W: The Media Converter ASR subsystem was unable to deliver hotword bargein event
246
CWVMC0011W: The Media Converter ASR subsystem was unable to deliver speech bargein event
CWVMC0012W: The Media Converter ASR subsystem received 0 RTP packets on port=VALUE_0 recognize mode=VALUE_1
Explanation: The WebSphere Voice Server did not receive any RTP packets from the MRCP client during the RECOGNIZE request for the specified session. The RECOGNIZE request was processed correctly. However, the RECOGNITIONCOMPLETE event that the MRCP client will receive may contain unexpected results since WebSphere Voice Server cannot process the audio stream for recognition. User response: Verify that the tuning parameters for your operating system are applied as described in the section on tuning operating systems in the WebSphere Voice Server Information Center.
CWVMC0013W: The Media Converter ASR subsystem did not accept an ASR media connection recognize mode=VALUE_0, number of audio bytes buffered=VALUE_1
CWVMC0014E: The Media Converter ASR subsystem encountered an IOException VALUE_0 packets received, number of audio bytes buffered=VALUE_1, number of audio bytes sent to ASR=VALUE_2
247
CWVMC0015E: The Media Converter ASR subsystem encountered an IOException=VALUE_0 while processing an ASR recognition match message
CWVMC0016E: The Media Converter ASR subsystem encountered a FileNotFoundException=VALUE_0

CWVMC0017E: The Media Converter ASR subsystem encountered an IOException=VALUE_0 while processing an ASR recognize request
CWVMC0018E: The Media Converter ASR subsystem encountered an IOException=VALUE_0 while processing a TTS end of prompt message
248
CWVMC0019E: The Media Converter ASR subsystem encountered an IOException=VALUE_0 while processing a TTS start of prompt message
CWVMC0020E: The Media Converter ControlSocket=VALUE_0 encountered an IOException=VALUE_1 while in the main run loop
CWVMC0021E: The Media Converter MediaSocket encountered an IOException=VALUE_0 while in the main run loop
CWVMC0022W: The Media Converter TTS subsystem encountered an IOException=VALUE_0 while processing an ASR bargein detected message
249
CWVMC0023W: The Media Converter TTS subsystem experienced a delay of VALUE_0 ms before receiving audio from the TTS engine.
CWVMC1000E: Unable to obtain a home object for the EJB using the JNDI name VALUE_0
Explanation: The Media Converter found an error when trying to access the specified JNDI name. User response: Make sure the WebSphere Voice Server enterprise application has been successfully deployed. Check the logs for more detail.
CWVMC1001E: Unable to create the VALUE_0 EJB instance

Explanation: The Media Converter found an error when trying to access the specified EJB instance. User response: Make sure the WebSphere Voice Server enterprise application has been successfully deployed. Check the logs for more detail.
CWVMD
CWVMD0001E: Failed to setup session for ASR - reason: VALUE_0
Explanation: WebSphere Voice Server could not establish a session for a MRCP Recognizer resource. User response: Check the WebSphere Voice Server Admin Console. Ensure the ASR engine status is green, and that there are available ASR engines. Check the logs for more detail.
CWVMD0002E: Failed to setup session for TTS - reason: VALUE_0

Explanation: WebSphere Voice Server could not establish a session for a MRCP Synthesizer resource. User response: Check the WebSphere Voice Server Admin Console. Ensure the TTS engine status is green, and that there are available TTS engines. Check the logs for more detail.
CWVMD0003E: Failed to process a MRCP Recognizer request- reason: VALUE_0

Explanation: Check the reason text for the cause of this error. User response: Check the WebSphere Voice Server Admin Console runtime and configuration panels. Ensure the runtime status of all components is green, and
250
that there are available engines. If the runtime status is green, check the logs for more detail.
CWVMD0004E: Failed to process synthesizer request- reason: VALUE_0

Explanation: Check the reason text for the cause of this error. User response: Check the WebSphere Voice Server Admin Console runtime and configuration panels. Ensure the runtime status of all components is green, and that there are available engines. If the runtime status is green, check the logs for more detail.
CWVMD0005E: Failed to send MRCP VALUE_0 event back to the MRCP client - reason: VALUE_1
Explanation: WebSphere Voice Server could not send the specified event back to the MRCP client. User response: Check the logs for more detail.
CWVMD0006E: Failed to teardown session for MRCP Recognizer reason: VALUE_0

Explanation: Recognizer resource reported an error during session teardown. User response: Check the logs for more detail.
CWVMD0007E: Failed to teardown session for MRCP Synthesizerreason: VALUE_0

Explanation: Synthesizer resource reported an error during session teardown. User response: Check the logs for more detail.
CWVMD0009E: Invalid MRCP active request id list was specified in the MRCP STOP request
Explanation: WebSphere Voice Server received a MRCP STOP request with an invalid list of request IDs. User response: This is an error in the MRCP client. Report the error to the MRCP client provider.
CWVMD0010E: Failed to process DEFINE-GRAMMAR request due to URI Syntax exception

Explanation: MRCP Server received a request containing an invalid grammar URI. User response: Ensure the client is sending a valid request.
CWVMD0011E: Failed to setup verification session. Reason: VALUE_0

Explanation: WebSphere Voice Server could not establish a session for a MRCP Verifier resource.
251
User response: Check the logs for more detail.
CWVMD0012E: Failed to teardown verification session. Reason: VALUE_0

Explanation: Verifier resource reported an error during session teardown. User response: Check the logs for more detail.
CWVMD0013E: Failed to process MRCP request targeting verifier resource. Reason: VALUE_0
Explanation: Check the reason text for the cause of this error. User response: Check the logs for more detail.
CWVMD1000E: Unable to reference the VALUE_0 Enterprise Java Bean

Explanation: MRCP Dispatcher encountered an error while trying to access its underlying resources. This problem is normally reported by the EJB container. User response: Ensure the WebSphere Voice Server installation completed successfully. Check the Installing section of the WebSphere Voice Server Information Center for more information.
CWVMD1001E: Unable to create the VALUE_0 Enterprise Java Bean instance

Explanation: MRCP Dispatcher found an error when trying to access its underlying resources. This problem is normally reported by the EJB container. User response: Ensure the WebSphere Voice Server installation completed successfully. Check the Installing section of the WebSphere Voice Server Information Center for more information.
CWVPT
CWVPT0000I: OK
Explanation: OK User response: No user action needed.
CWVPT0001I: voiceTest completed VALUE_0 command(s) successfully

Explanation: Success User response: No user action needed.
CWVPT0002E: Error reading file VALUE_0

Explanation: An error occurred while reading the file. User response: Check the file.
252
CWVPT0003E: Cannot read file VALUE_0

Explanation: The file could not be opened for reading. User response: Check the file name, access permission, etc.
CWVPT0004E: Option/Command not expected: VALUE_0

Explanation: The command line could not be parsed. User response: Correct the command line or the command file.
CWVPT0005E: Option not expected: VALUE_0

Explanation: An unknown option or an option that does not apply to the context has been found. User response: Correct the command line or the command file.
CWVPT0006E: Option VALUE_0 is already set (VALUE_1)

Explanation: Multiple values cannot be set with this option. User response: Correct the command line or the command file.
CWVPT0007E: Cannot write file VALUE_0

Explanation: The file could not be opened for writing. User response: Check the file name, access permission, etc.
CWVPT0008E: File format is not supported: VALUE_0

Explanation: The audio file is in a format that is not understood. User response: Convert the file to either .wav or .au with mu-law or A-law content.
CWVPT0009E: Audio must be one channel - 8kHz, A-law or mu-law (VALUE_0)

Explanation: The audio format is not supported. User response: Convert the file to either .wav or .au with mu-law or A-law content.
CWVPT0010E: RTSP error VALUE_0 VALUE_1

Explanation: The RTSP server returned an error. User response: Check the server.
253
CWVPT0011E: Could not parse RTSP: VALUE_0

Explanation: The reply from the RTSP server is not following the protocol. User response: Check the server and contact support.
CWVPT0012E: Malformed RTSP header: VALUE_0

Explanation: The reply from the RTSP server is not following the protocol. User response: Check the server and contact support.
CWVPT0013E: Could not find server port in Transport header: VALUE_0

Explanation: The Transport header does not contain an RTP server endpoint. User response: Check the server and contact support.
CWVPT0014I: Sending teardown

Explanation: Information User response: No user action needed.
CWVPT0015E: Received bad RTSP content

Explanation: The RTSP reply or query has content of an unknown type. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVPT0016E: MrcpConnection mode must be play or record

Explanation: Invalid parameter User response: Correct the parameter value to either play or record.
CWVPT0017E: URL must start with rtsp:// (VALUE_0)

Explanation: Only RTSP URLs are accepted. User response: Use an RTSP URL.
CWVPT0018E: Malformed URL: VALUE_0

Explanation: The URL could not be parsed or has an unsupported protocol. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/
254
CWVPT0019E: Unknown host: VALUE_0

Explanation: The host could not be contacted. User response: Check the host name or the IP address.
CWVPT0020E: Connection refused by host VALUE_0 on port VALUE_1

Explanation: The host did not accept the connection. User response: Check the host name (or IP address) and port. Ensure the server is running.
CWVPT0021E: Failed to open IO stream for RTSP socket (VALUE_0 VALUE_1)

Explanation: A socket connection error occurred. User response: Check the network, contact support.
CWVPT0022E: Error reading from RTSP socket

CWVPT0023E: Error writing into RTSP socket

CWVPT0024E: Port out of range (VALUE_0)

Explanation: Invalid port, it must be > 1024 User response: Correct the parameter value.
CWVPT0025E: Cannot bind socket to port VALUE_0

Explanation: Another application has already exclusively bound a socket on this port. User response: Use a different port value.
CWVPT0026E: Security exception while binding RTP datagram socket

Explanation: Java security exception User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/
255
CWVPT0027I: Connecting to VALUE_0 VALUE_1

CWVPT0028I: RTP contact at VALUE_0 VALUE_1

CWVPT0029I: Waiting for reply

Explanation: The client is waiting for reply of completion event. User response: No user action needed.
CWVPT0030I: RTSP session: VALUE_0

CWVPT0031E: Could not retrieve grammar URL

Explanation: The grammar URL has not been delivered by the MRCP server. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVPT0032E: Expected HTTP URL (VALUE_0)

Explanation: Only HTTP URLs are expected here User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVPT0033E: Cannot open connection to fetch URL (VALUE_0)

Explanation: The HTTP server is not running or it has refused the connection User response: Check the HTTP server and contact support.
256
CWVPT0034E: Error reading URL stream (VALUE_0)

Explanation: An error occurred while downloading from the URL User response: Check the server and contact support.
CWVPT0035E: -audioTTS is required with synthesize

Explanation: Required option User response: Check the command line or file.
CWVPT0036E: -speak is required with synthesize

CWVPT0037E: -audioASR is required with recognize

CWVPT0038E: -grammar is required with define

CWVPT0039E: setup(...) must be called before

Explanation: The connection is not in the right state. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVPT0040E: -sdpPcm must be mulaw or alaw (VALUE_0)

Explanation: Invalid option value User response: Correct the command line.
CWVPT0041I: Received start of speech after VALUE_0 ms

257
CWVPT0042I: Time to complete: VALUE_0 ms

CWVPT0043E: Timed out during SETUP (VALUE_0 ms)

Explanation: The server did not respond to a SETUP request within the timeout User response: Increase the timeout and check the server.
CWVPT0044E: Timed out in MRCP request

Explanation: The server did not reply to an MRCP request within the timeout User response: Increase the timeout and check the server.
CWVPT0045E: Invalid frame duration (VALUE_0), must be at least 20 ms

Explanation: Invalid option value. User response: Correct the command line.
CWVPT0046E: Circular command file reference (VALUE_0)

Explanation: A command file has references to itself. User response: Remove the circular reference.
CWVPT0047W: Commands or options follow -commandFile option, will be ignored

Explanation: If present option -commandFile must end the command line or the file being included. User response: Remove the extraneous commands or options.
CWVPT0048E: No supported audio files found in this directory (VALUE_0)

Explanation: The directory is empty or doesnt have any audio files in the supported formats. User response: Check the command line options.
CWVPT0049E: A value is expected this for option (VALUE_0)

Explanation: A value must be given to this optional parameter. User response: Check the command line options.
258
CWVPT0050E: Invalid value (VALUE_0) for option VALUE_1, expected VALUE_2

Explanation: Invalid option value User response: Correct option value to one of the expected values.
CWVPT0051I: Reading file VALUE_0

Explanation: Attempting to read the file. User response: No action needed.
CWVPT0052I: Writing file VALUE_0

Explanation: Attempting to write file. User response: No action needed.
CWVPT0053E: Invalid RTP server port (VALUE_0) in Transport header: VALUE_1

Explanation: The Transport contains an invalid RTP server port. User response: Contact support.
CWVPT0054E: A value is needed for this header: VALUE_0

Explanation: This header must be given a value. User response: Contact support.
CWVPT0055I: Elapsed time from the start of audio: VALUE_0 ms, from the end of the audio: VALUE_1 ms
CWVPT0099I: <<VALUE_0>>
Explanation: Trace message User response: No user action needed.
CWVRB
CWVRB0001E: Illegal value for parameter VALUE_0: VALUE_1
Explanation: The value provided for the given parameter is invalid. User response: If the invalid parameter is from WebSphere Voice Server, refer to the Information Center for the range of valid values.
CWVRB0002E: Unsupported parameter: VALUE_0

Explanation: The parameter provided in this MRCP message is not supported.
259
CWVRB0003E: Configuration error: VALUE_0=VALUE_1 - using default: VALUE_2

Explanation: A configuration error has occurred. User response: Please verify if installation was successful and that the value specified conforms to the range of valid values in the Information Center.
CWVRB0004E: DEFINE-GRAMMAR is not allowed when Recognizer is in RECOGNIZING state

Explanation: The recognizer must be stopped (in IDLE or RECOGNIZED states) in order to be able to process this request. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0005I: GET-RESULT is not supported

Explanation: The GET-RESULT MRCP message is not supported by WebSphere Voice Server in this version. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0006E: Error fetching grammar: VALUE_0

Explanation: An error occurred while attempting to fetch this grammar. User response: Ensure that this grammar resource is accessible.
CWVRB0007E: Grammar contains an unsupported format. grammar:VALUE_0

Explanation: The given grammar has an unsupported format. User response: Use only supported grammar formats as specified in the Information Center.
260
CWVRB0008E: Syntax or semantic error in grammar:VALUE_0

Explanation: The given grammar contains a syntax or semantic error. User response: Correct this grammar.
CWVRB0009E: Grammar error: VALUE_0

Explanation: An error occurred with this grammar. See SystemOut.log for more details. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0010E: RECOGNIZE is not allowed when Recognizer is in RECOGNIZING state.

Explanation: Server received a RECOGNIZE request while another one was already in progress. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0011E: Grammar has not been defined: VALUE_0

Explanation: The grammar, referenced during a RECOGNIZE request, was not successfully defined to WebSphere Voice Server by the speech application. User response: This message indicates a problem with the Interactive Voice Response (IVR) system. Contact support for the IVR system or application you are using.
CWVRB0012E: Pool Exception: VALUE_0

Explanation: An error occurred while accessing the pool of recognizer resources. See message for more information. User response: Check log for additional information. You may need to increase the number of configured ASR engines to start.
CWVRB0013E: Error sending RECOGNITION-COMPLETE event to SpeechListenerMgr object: VALUE_0

Explanation: An error occurred while attempting to send a RECOGNITIONCOMPLETE event to the Speech Listener EJB. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the
261
CWVRB0014E: ASR grammar pre-loads error: VALUE_0

Explanation: Failed to pre-load grammars. See message for more details. User response: Correct the problem that prevents the grammar from being loaded.
CWVRB0015E: ASR setup error for locales VALUE_0: VALUE_1. ASR engine allocation interrupted
Explanation: An error occurred while attempting to set up the engines. User response: Ensure that the installation was successful for all languages and the system has enough hardware resources to run this configuration.
CWVRB0016E: ASR setup error: VALUE_0

Explanation: Error occured while setting up ASR. See message for more details. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0018E: Error: Configuration property VALUE_0 has not been set

Explanation: Configuration error occurred. See message for more information. User response: Check log files for additional information.
CWVRB0019E: Error validating default configuration

CWVRB0020E: Recognizer setup must be invoked prior to VALUE_0. Reason: null context
Explanation: MRCP SET-UP for Recognizer must be received prior to other requests.
262
CWVRB0022E: Failed to release grammar VALUE_0- reason VALUE_1

Explanation: Request to undefine a grammar from the cache subsystem failed. This situation should not cause any interruption of service but it will impact the systems ability to remove expired grammars from the cache subsystem. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0023W: The GrammarUsage object was not found for VALUE_0

Explanation: Grammar utilization information was not found for the given grammar. This situation should not cause any interruption of service but it will impact the systems ability to remove expired grammars from the cache subsystem. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0024E: Error sending START-OF-SPEECH event to SpeechListenerMgr object: VALUE_0

Explanation: An error occurred while attempting to send this event to the Speech Listener EJB. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0025E: Preload grammar failed: VALUE_0 reason: VALUE_1

Explanation: An error occurred while processing this grammar. User response: Verify the given grammar is correctly entered into the configuration. Ensure the given grammar has no syntax errors.
CWVRB0026W: Charset VALUE_0 is not supported

Explanation: The provided charset is not supported.
263
CWVRB0027E: Factory error: wvs.runtime.mode=VALUE_0 J2EE=VALUE_1 Exception=VALUE_2

Explanation: Internal error initializing the Recognizer Factory. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0028E: Specified URI VALUE_0 has an invalid syntax

Explanation: The provided string is not a compliant URI string. User response: This error could have been caused by entering an invalid URI in a list of pre-defined grammars or in a VoiceXML application.
CWVRB0029E: Specifying multiple grammar languages in one RECOGNIZE request is not supported -languages detected: VALUE_0, VALUE_1
Explanation: RECOGNIZE request must contain grammars for a single language. User response: Correct the application.
CWVRB0030E: Error returning DTMF processor to the pool: VALUE_0 VALUE_1

Explanation: Failed to release DTMF resources from the current session. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0031E: Error starting failover thread to recover ASR engines. Reason: VALUE_0
Explanation: Error starting a background thread to recover the system. System may be running low on resources. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/
264
CWVRB0032E: Error trying to recover ASR Bean VALUE_0

CWVRB0033E: Configuration error: no valid ASR languages were installed

CWVRB0034E: Internal error: invalid ID object:VALUE_0

CWVRB0035E: Error processing recognition results: Error=VALUE_0: IPEvent=VALUE_1

Explanation: Internal error occurred while processing the recognition results. User response: Ensure that application code, including grammars and their semantic code has no errors.
CWVRB0036E: The InputProcessor object encountered an error: VALUE_0: VALUE_1

Explanation: A recognizer resource was unable to carry out an action requested by the recognizer component. The error message contains some detail about the failure, However, the cause cannot be immediately determined.
265
CWVRB0037E: Language exception for grammar VALUE_0. Reason: VALUE_1

Explanation: Language exception occurred while compiling this grammar. User response: See reason text for additional information.
CWVRB0038E: Error retrieving Recognizer-Context-Block from ASR resource: VALUE_0

CWVRB0039E: RECOGNITION-START-TIMERS failed. Mode: VALUE_0. Reason: VALUE_1

Explanation: Request to start no-input timers failed. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0040E: Error stopping input processor: VALUE_0 - reason: VALUE_1

Explanation: An error occurred while stopping processor. User response: Check log for additional information.
CWVRB0041I: ASRBean added to failover queue -details: VALUE_0

266
CWVRB0042I: ASRBean failover succeeded - details: VALUE_0

CWVRB0043E: Installation error: problem validating installed locale: VALUE_0

Explanation: The installation program contains an invalid locale. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0044E: SET-PARAMS is not allowed in this state: VALUE_0

Explanation: The recognizer is in a processing state that will not accept this request. The request may be out of sequence because either another error occurred in WebSphere Voice Server or because there are problems with the MRCP client. The recognizer must be in either the IDLE or RECOGNIZED processing state to accept this request. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0045E: SET-PARAMS: Recognizer-Context-Block: missing parameters: content-id=VALUE_0, context-block size:VALUE_1 bytes

Explanation: When setting Recognizer-Context-Block, the MRCP request must contain the content-id of the resource as well as the context-block. User response: Application must be fixed.
CWVRB0046E: SET-PARAMS: Recognizer-Context-Block: Error: VALUE_0

Explanation: Error setting Recognizer-Context-Block.
267
CWVRB0047E: RECOGNIZE request failed: VALUE_0

Explanation: An error occurred during a RECOGNIZE request. The request could not be completed. The error message contains some detail about the failure that occurred. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRB0048W: ASR pool size is VALUE_0- no ASR engines will be available

Explanation: No ASR engines will be available for the provided installed languages. User response: Change the configuration to have ASR engines started.
CWVRB0049I: Recognizer initialization for VALUE_0 is complete. VALUE_1 out of VALUE_2 configured ASR engine(s) available
Explanation: Informs the initialization status of the Recognizer. User response: None
CWVRB0050W: WARNING: Received NO-INPUT from same source that sent INPUT-STARTED!
CWVRB0051E: Grammars must be provided with MRCP RECOGNIZE request

268
CWVRB1000E: Unable to obtain a home object for the EJB using the JNDI name VALUE_0
CWVRB1001E: Unable to create the VALUE_0 EJB instance

CWVRB1002E: An unexpected runtime exception was caught during VALUE_0 processing

CWVRR 0000 - 4999

CWVRR0000E: Resource VALUE_0 has invalid resource type of Serializable
269
CWVRR0001E: Resource VALUE_0 has invalid resource type of Stream

CWVRR0002E: Resource VALUE_0 has unknown resource type

CWVRR 5000 - 9999

CWVRR5000E: Application Resource Fetch Error: VALUE_0
Explanation: The application resource could not be fetched from the web or application server. User response: Use a Visual browser to ensure the resource is available for the URI being requested.
CWVRR5001W: Application Resource Fetch Timeout: VALUE_0

Explanation: The resource could not be fetched in the time specified by the fetch time out value. User response: Ensure that the fetch time out value is appropriate.
CWVRR5002I: ResResolver configuration : VALUE_0

Explanation: Configuration information related to the http response cache. User response: Adjust public settings if necessary.
CWVRR5003E: End of fetch due to file not found VALUE_0

Explanation: The resource could not be located on the file system. User response: Ensure the resource is correctly specified by the URI.
270
CWVRR5004E: End of fetch due to error: VALUE_0 : System Message: VALUE_1

CWVRR5005W: Invalid configuration parameter value: VALUE_0 : VALUE_1

Explanation: The configuration value is illegal or out of range. User response: Please verify the configuration value range and modify the property as necessary.
CWVRR6000E: Cache Manager is not open

CWVRR6001E: File does not exist

CWVRR6002E: Cannot delete file

271
CWVRR6003W: Memory cache capacity exceeded: VALUE_0 : VALUE_1

Explanation: The resource cannot be placed in the memory cache because its size exceeds the memory cache capacity. The resource will not be cached. User response: Increase the memory cache size to accomodate large resources.
CWVRT
CWVRT0001W: Unable to validate RTSP configuration property VALUE_0 - defaulting to port VALUE_1
Explanation: There was either a problem loading the WebSphere Voice Server configuration file(s), or the value was not properly specified. User response: Make sure the configuration file(s) exist and/or they are in the correct directory, and make sure the value specified is correct.
CWVRT0002W: Unable to validate RTSP configuration property VALUE_0 - defaulting server timeout to VALUE_1 seconds
CWVRT0003W: Unable to validate RTSP configuration property VALUE_0- defaulting socket timeout to VALUE_1 seconds
CWVRT0004W: Unable to validate RTSP configuration property VALUE_0 - defaulting reco media url to VALUE_1
CWVRT0005W: Unable to validate RTSP configuration property VALUE_0 - defaulting synth media url to VALUE_1
272
CWVRT0006I: The RTSP Bridge is started

Explanation: The RTSP Bridge is started. User response: No action required.
CWVRT0007I: The RTSP Bridge is shutdown

Explanation: The RTSP Bridge is shutdown. User response: No action required.
CWVRT0009W: The RTSP Bridge has already been initialized

Explanation: This will happen if you try to call the init() method more than once. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRT0010W: The RTSP Bridge has already been destroyed.

Explanation: This will happen if you try to call the destroy() method more than once. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRT0011E: MRCP Client was unable to provide a read/write stream for the WebSphere Voice Server - additional information follows: MRCP Client: VALUE_0 Error Message: VALUE_1
Explanation: This will happen when a socket connection is made and the RTSP Bridge is unable to get the input and output stream from the socket. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRT0012W: Unable to validate RTSP configuration property VALUE_0 - defaulting to Boolean value VALUE_1
Explanation: There was either a problem loading the WebSphere Voice Server configuration file(s), or the value was not properly specified.
273
User response: Make sure the configuration file(s) exist and/or they are in the correct directory, and make sure the value specified is correct.
CWVRT0013W: Time out has occured on the session - no more requests were available from the client within the specified time limit. Additional information follows: MRCP Client: VALUE_0 Error Message: VALUE_1
Explanation: This will happen if the session thread is blocked reading the input stream for longer than the specified socket timeout (meaning there is no data to read from the socket) User response: Try increasing the socket timeout configuration parameter (com.ibm.voice.server.rtspbridge.sockettimeout) to a larger value.
CWVRT0014W: The MRCP Client has unexpectedly disconnected from an RTSP Session during a read/write operation with the WebSphere Voice Server. Additional information follows: MRCP Client: VALUE_0 Error Message: VALUE_1
Explanation: The primary socket that was established by the client for communicating with WebSphere Voice Server has been inadvertently closed by the client, causing a read/write error on WebSphere Voice Server. User response: This message indicates a problem with the Interactive Voice Response (IVR) system. Contact support for the IVR system or application you are using.
CWVRT0015W: The MRCP Client sent a malformed RTSP message to the WebSphere Voice Server. Additional information follows: MRCP Client: VALUE_0 Error Message: VALUE_1
Explanation: The RTSP message is not formatted as outlined by the HTTP 1.1 specification. All HTTP-like messages, like RTSP, must follow a strict format in order to be parsed correctly. This format is defined in the HTTP 1.1 specification and described in RFC 2326 - Real Time Streaming Protocol (RTSP). User response: This message indicates a problem with the Interactive Voice Response (IVR) system. Contact support for the IVR system or application you are using.
CWVRT0016W: The MRCP Client sent an invalid RTSP message to the WebSphere Voice Server. Additional information follows: MRCP Client: VALUE_0 Error Message: VALUE_1
Explanation: There was a problem with the RTSP message. For example, the session ID does not match the current session ID. Or, the RTSP version is not correct. User response: Typically the RTSP response code will be very clear as to why the message could not be validated. Use a packet sniffer like Ethereal, to investigate the response code.
274
CWVRT0017E: An internal WebSphere Voice Server error was detected by the RTSP Bridge component - additional information follows: MRCP Client: VALUE_0 Error Message: VALUE_1
Explanation: There was an internal problem with the RTSPBridge. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRT0018W: An internal WebSphere Voice Server error was detected by the RTSPBridge component - additional information follows: MRCP Client: VALUE_0 Error Message: VALUE_1
Explanation: There was an internal problem with the RTSPBridge. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRT0020W: The MRCP Client acknowledged an event from the WebSphere Voice Server with an invalid sequence number: VALUE_0 MRCP Client: VALUE_1
Explanation: This error will occur if a client initiated response cannot be validated by the WebSphere Voice Server. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRT0021W: The MRCP Client acknowledged an event from the WebSphere Voice Server with an unexpected response: Invalid response code: VALUE_0 MRCP Client: VALUE_1
Explanation: Could not process the server initiated request (This warning will occur if the client initiated response is not a 200 Ok). User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
275
CWVRT0022E: The RTSP Server was unable to close all the session threads before continuing shutdown. There are VALUE_0 left.
Explanation: During an immediate shutdown of the RTSP Server, all session threads are expected to terminate before shutting down remaining server resources. If that is not the case, this error will be generated. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRT1000E: Unable to obtain a home object for the EJB using the JNDI name VALUE_0
Explanation: There was a problem looking up the EJB User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVRT1001E: Unable to create the VALUE_0 EJB instance

Explanation: There was a problem creating the EJB instance User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVSB
CWVSB0001E: Failed to launch TTS engine process. Reason: VALUE_0
Explanation: Check the reason text for the cause of this error. User response: Check for invalid configuration settings. See the Information Center Installation Troubleshooting Tips for more information.
CWVSB0002E: Failed to terminate TTS engine process. Reason: VALUE_0

276
CWVSB0003E: Failed to setup Synthesizer session. Reason VALUE_0

Explanation: Synthesizer found a problem when establishing a session. User response: Check the Admin console and make sure the Synthesizer component is up and running. Check if you have enough TTS engines available to establish a new session.
CWVSB0004E: MRCP VALUE_0 request has failed. Reason: VALUE_1

Explanation: Check the reason text for the cause of this error. User response: Check the logs for more detail.
CWVSB0005E: Connection with TTS engine has been closed - the WebSphere Voice Server is starting a new TTS engine
Explanation: The MRCP Synthesizer lost a connection with a TTS engine and a new process is being started. User response: Check the logs for more detail.
CWVSB0006W: A default TTS voice has not been configured - using default voice VALUE_0
Explanation: A default TTS voice has not been configured. User response: Configure a default TTS voice using the WebSphere Voice Server Administrative Console configuration panel.
CWVSB0007E: Missing required configuration for Synthesizer - ensure the property VALUE_0 is configured
Explanation: Missing the specified configuration property. User response: Check the configuration and make sure the specified property is specified.
CWVSB0009W: Invalid value for property VALUE_0 - setting default value

Explanation: Invalid value for the specified configuration property. User response: Make sure the specified property has a valid value.
CWVSB0010I: Starting Synthesizer...

Explanation: Informational message indicating that the Synthesizer startup has started. User response: None.
277
CWVSB0011I: Synthesizer startup complete - status is VALUE_0 VALUE_1 out of VALUE_2 configured TTS engine(s) available
Explanation: Informational message indicating that the Synthesizer has started. User response: Check the status and make sure all configured TTS engines are available.
CWVSB0012I: Shutting down Synthesizer...

Explanation: Informational message indicating that the Synthesizer is stopping. User response: None.
CWVSB0013I: Synthesizer shutdown complete

Explanation: Informational message indicating that the synthesizer has been stopped. User response: None.
CWVSB0014W: Synthesizer pool size is 0. No TTS engines will be available

Explanation: Indicates that there are no TTS engines available, since none has been configured to start. User response: None.
CWVSB0015E: Invalid configuration parameter. Parameter: VALUE_0, Value: VALUE_1

Explanation: Invalid value for the specified configuration property. User response: Make sure the specified property has a valid value.
CWVSB0016E: Failed to retrieve configuration property: VALUE_0

Explanation: Missing the specified configuration property. User response: Make sure you have the specified property defined.
CWVSB0018E: Failed to schedule TTS failover thread

Explanation: Synthesizer has lost connection with a TTS engine but its failover mechanism failed. At least one of the configured TTS engines is not available to process any further requests. User response: Check the logs for more detail.
CWVSB0019I: FAILOVER: Starting a new TTS engine - VALUE_0 out of VALUE_1 attempts
Explanation: Informational message indicating that a new TTS engine is being started.
278
User response: This message indicates that the Synthesizer has lost connection to one of the TTS engines. Check the log for more detail.
CWVSB0020I: FAILOVER: New engine has been started successfully pool size: VALUE_0
Explanation: Informational message indicating that a new TTS engine has been started. User response: This message indicates that the Synthesizer has lost connection to one of the TTS engines. Check the log for more detail.
CWVSB0021E: FAILOVER: Failed to restart TTS engine

Explanation: Synthesizer failover mechanism could not launch a new TTS engine. User response: Check the logs for more detail.
CWVSB0022E: FAILOVER: Failover process failed - Synthesizer is partially operational

Explanation: Synthesizer failover mechanism failed to start a new TTS engine. User response: Check if Synthesizer component still have engines available. Check the logs for more details. Restart the server if necessary.
CWVSB0023W: Failed to load TTS lexicon file(s): VALUE_0

Explanation: The specified lexicon file could not be loaded. User response: Check the configuration. Make sure the lexicon URI is accessible.
CWVSB0024E: Invalid attribute value VALUE_0,VALUE_1

Explanation: The synthesizer could not validate the specified attribute. User response: Make sure the specified attribute is being specified correctly.
CWVSB0025E: Synthesizer has failed to initialize VALUE_0 engines the startup process is being aborted
Explanation: Synthesizer startup process failed because one or more engines could not be launched. User response: Make sure TTS is correctly installed. Check configuration settings. Check the logs for more detail.
CWVSB0026W: Failed to send trace message to the TTS engine - TTS engine trace may not work correctly
Explanation: The TTS engine trace could not be set. Trace messages will not be available. User response: Make sure the TTS trace is configured correctly.
279
CWVSB0027E: Failed to tear down Synthesizer session - reason VALUE_0

Explanation: Synthesizer found problems during session teardown. User response: Check the logs for more detail.
CWVSB0028W: Unable to fetch the following lexicon file(s): VALUE_0

Explanation: The TTS engine failed to fetch the specified lexicon file. User response: Make sure the lexicon URI is correct and the specified resource is available.
CWVSB0029W: Error parsing the following lexicon file(s): VALUE_0

Explanation: The TTS engine failed to parse the specified lexicon file. User response: Check the lexicon syntax. Enable TTS engine trace and try again for further detail.
CWVSB0030E: Failed to set parameters for MRCP SPEAK request. Error code: VALUE_0
Explanation: The TTS engine replied with an error when validating the SPEAK request parameters. User response: Check the logs for more detail.
CWVSB0031W: Failed to delete segment info file: VALUE_0

Explanation: Synthesizer failed to delete the specified file. User response: Check the logs for more detail.
CWVSB0032E: Synthesizer configuration error - default voice VALUE_0 does not match any of the available voices
Explanation: The Synthesizer requires the default voice to be selected from the list of available voices. User response: Check the configuration and make sure that the configured default voice matches one of the available voices.
CWVSB1000E: Unable to obtain a home object for the EJB using the JNDI name VALUE_0
Explanation: Synthesizer found an error when trying to access the specified JNDI name. User response: Make sure the WebSphere Voice Server enterprise application has been successfully deployed. Check the logs for more detail.
280
CWVSB1001E: Unable to create the VALUE_0 EJB instance

Explanation: Synthesizer found an error when trying to access the specified EJB instance. User response: Make sure the WebSphere Voice Server enterprise application has been successfully deployed. Check the logs for more detail.
CWVSM
CWVSM0003W: The configuration property VALUE_0 was not found defaulting to value VALUE_1
Explanation: The server attempted to use a configuration property that does not exist. User response: Either the configuration is damaged, or the server was unable to load the configuation information.
CWVSM0004E: After VALUE_0 attempts - the server was unable to find an available port to create an RTP audio input server. The Last port attempted was VALUE_1.
Explanation: If the system is overloaded, resources such as ports may not be available. User response: Make sure the system is operating within an appropriate load, then retry the operation.
CWVSY 1000 - 2999, 9000 - 9999

CWVSY1000E: No command line parameters passed.
CWVSY1001E: WebSphere Voice Server installation verification failed.

Explanation: The WebSphere Voice Server installation verification test has found a problem during either execution or validation. User response: You should review the system log file for the specific error. For additional information on resolving this error, please consult the IBM WebSphere Voice Server Support Website available at: http://www.ibm.com/software/ pervasive/voice_server/support/index.html. The site provides searchable databases of technotes, solutions, and e-fixes, as well as contact information for the WebSphere Support team.
281
CWVSY1002I: WebSphere Voice Server installation verification has started.

Explanation: The WebSphere Voice Server installation verification test is starting its processing. User response: No user action needed.
CWVSY1003I: WebSphere Voice Server installation verification executing for VALUE_0 ...
Explanation: The WebSphere Voice Server installation verification test is executing for V ALUE_0 (for either ASR or TTS). User response: No user action needed.
CWVSY1004I: WebSphere Voice Server installation verification worked successfully.

Explanation: The WebSphere Voice Server installation verification test worked correctly. User response: No user action needed.
CWVSY1005I: WebSphere Voice Server installation verification is validating results ...

Explanation: The WebSphere Voice Server installation verification test is validating the results. User response: No user action needed.
CWVSY1006I: WebSphere Voice Server installation verification is complete.

Explanation: The WebSphere Voice Server installation verification test has finished. User response: No user action needed.
CWVSY1007E: WebSphere Voice Server installation verification provided invalid parameter: VALUE_0.
282
CWVSY1008I: WebSphere Voice Server installation verification successfully created a sample audio file, named VALUE_0 and located in the $WVS_ROOT/ivt directory.
Explanation: The TTS voice has been installed correctly. The audio file was created in the $WVS_ROOT/ivt directory. User response: You may play the sample audio file in the $WVS_ROOT/ivt directory.
CWVSY2000E: Unable to create the WebSphere Voice Server logging directory VALUE_0. The exception message is VALUE_1.
CWVSY2001I: WebSphere Voice Server runtime threading service started

Explanation: WebSphere Voice Server has started the runtime threading service successfully. User response: No user action needed.
CWVSY2002I: Starting WebSphere Voice Server

Explanation: WebSphere Voice Server is initializing. User response: No user action needed.
CWVSY2003I: WebSphere Voice Server started

Explanation: WebSphere Voice Server started successfully. User response: No user action needed.
CWVSY2004I: Stopping WebSphere Voice Server

Explanation: WebSphere Voice Server is shutting down and stopping all engines. User response: No user action needed.
CWVSY2005I: WebSphere Voice Server stopped

Explanation: WebSphere Voice Server has stopped. User response: No user action needed.
283
CWVSY2006E: WebSphere Voice Server failed to start - see the SystemOut.log file for more information
Explanation: WebSphere Voice Server encountered an error during its initialization. User response: You should review the SystemOut.log file for the specific error. For additional information on resolving this error, please consult the IBM WebSphere Voice Server Support Website available at: http://www.ibm.com/software/ pervasive/voice_server/support/index.html. The site provides searchable databases of technotes, solutions, and e-fixes, as well as contact information for the WebSphere Support team.
CWVSY2007E: WebSphere Voice Server encountered a failure during the stop operation - see the SystemOut.log file for more information
Explanation: WebSphere Voice Server encountered an error during the stop operation. User response: You should review the SystemOut.log file for the specific error. For additional information on resolving this error, please consult the IBM WebSphere Voice Server Support Website available at: http://www.ibm.com/software/ pervasive/voice_server/support/index.html. The site provides searchable databases of technotes, solutions, and e-fixes, as well as contact information for the WebSphere Support team.
CWVSY2008E: An exception has occurred. The exception message is: VALUE_0

CWVSY2009I: Loading WebSphere Voice Server configuration

Explanation: WebSphere Voice Server is in the process of loading its configuration information. User response: No user action needed.
CWVSY2010I: Loaded WebSphere Voice Server configuration

Explanation: WebSphere Voice Server has completed loading its configuration information. User response: No user action needed.
284
CWVSY2011E: WebSphere Voice Server failed to load the configuration information - see the SystemOut.log file for more information
Explanation: WebSphere Voice server failed to load the configuration information needed to start. User response: You should review the system log file for the specific error. For additional information on resolving this error, please consult the IBM WebSphere Voice Server Support Website available at: http://www.ibm.com/software/ pervasive/voice_server/support/index.html. The site provides searchable databases of technotes, solutions, and e-fixes, as well as contact information for the WebSphere Support team.
CWVSY2012I: VALUE_0, build date: VALUE_1

Explanation: WebSphere Voice Server product level V ALUE_0 and build date V ALUE_1 information is logged. User response: No user action needed.
CWVSY2013E: WebSphere Voice Server cannot obtain MANIFEST.MF information

Explanation: WebSphere Voice Server product level and build date information could not be obtained from the MANIFEST.MF (part of the wvsapp.ear file), located in the $WVS_ROOT/ear directory. User response: No user action needed.
CWVSY9905I: %s file is successfully opened

Explanation: Native logging subsystem successfully located and opened WebSphere Voice Server configuration repository file. User response: No user action needed.
CWVSY9910E: property file not found: %s

Explanation: Native logging subsystem was unable to locate WebSphere Voice Server configuration repository file. User response: There is probably a problem with missing environment variables or installation configuration.
CWVSY9912E: ERROR: could not initialize logger myLog

Explanation: Native logging subsystem cannot initialize logging runtime for native components. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
285
CWVSY9915E: sock open error %d.

Explanation: Native logging subsystem was unable to open UDP socket connection to Java logging service. The UDP port number associated with logging subsystem might be allocated to another service in system. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVSY9920E: sock connect error: %d.

CWVSY9922E: sock close error: %d.

Explanation: Native logging subsystem was unable to close UDP socket connection to Java logging service. User response: No user action needed.
CWVSY9925E: send error %d.

Explanation: Native logging subsystem was unable to send log or trace date to Java logging service. There could be a problem with WebSphere Voice Server logging services or UDP socket is unavailable. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/ index.html. The site also provides searchable databases of technotes, solutions, and e-fixes.
CWVSY9930E: message send error=%d.

Explanation: Native logging subsystem was unable to send log or trace date to Java logging service. There could be a problem with WebSphere Voice Server logging services or UDP socket is unavailable. User response: Refer to the troubleshooting section of the WebSphere Voice Server information center for help diagnosing the problem. If necessary, submit the problem through the IBM WebSphere Voice Server Support Web site, which is available at: http://www.ibm.com/software/pervasive/voice_server/support/
286
CWVSY9935E: cannot get the port number from property file

Explanation: Native logging subsystem was unable to get UDP port number configuration value from WebSphere Voice Server configuration repository. User response: There is probably a problem with missing environment variables or installation configuration.
CWVSY9940I: port 65000 is used

Explanation: Native logging subsystem defaulted to use UDP port 65000 for communication with Java logging service. User response: No user action needed.
CWVSY9942E: cannot get the port number from .cwvs.properties need com.ibm.voice.server.common.log.logservice-udpport=xxxx
Explanation: Native logging subsystem was unable to get UDP port number configuration value from WebSphere Voice Server configuration repository. User response: There is probably a problem with missing environment variables or installation configuration.
CWVSY9943I: port 65000 is used

Explanation: Native logging subsystem defaulted to use UDP port 65000 for communication with Java logging service. User response: No user action needed.
CWVSY9945E: open socket error: %d

CWVSY 3000 - 3999

CWVSY3000E: WebSphere Voice Server configuration MBean was unable to load the configuration properties
287
CWVSY3001E: WebSphere Voice Server property name VALUE_0 does not exist in the product configuration
CWVSY3002E: Specified filter parameter of null is not valid for a configuration property search
CWVSY3003E: Read operation of the WebSphere Voice Server configuration file VALUE_0 failed
CWVSY3004E: Write operation of the WebSphere Voice Server configuration file VALUE_0 failed
288
CWVSY3005E: WebSphere Voice Server configuration information failed to load

CWVTE
CWVTE0001E: unable to write ESPR to file: eXsXpXr.dbg errno = %s
CWVTE0002W: %s is not supported

Explanation: Unsupported language. User response: Check your configuration and if correct, install missing language voices.
CWVTE0003W: language=%2$s, dialect=%3$s not supported

Explanation: Unsupported language and dialect combination. User response: Check your configuration and if correct, install missing language voices.
CWVTE0004W: Could not load CMM library

Explanation: Cannot find or read CMM shared library. User response: Check the location and permissions of libibmcmm.so (Linux) / ibmcmm.dll (Windows) / libibmcmm.a (AIX).
CWVTE0005W: No CMM library registered

Explanation: ibmcmm.dll not registered User response: Reinstall
289
CWVTE0006E: hDllInstance=%s, fdwReason=ERROR

CWVTE0007W: Invalid parameter - returning %s

CWVTE0008W: returning: %s
CWVTE0009W: failed: returning: %s

CWVTE0010W: eciReset failed - retval=%s

290
CWVTE0011W: failed - returning %s





291




292





293





294




295
CWVTE0034W: returning %s




296





297




298




CWVTE0052W: Reentered from the same thread

299
CWVTE0053W: malloc failed - returning NULL_ECI_HAND

CWVTE0054W: CreateWindow failed - returning NULL_ECI_HAND

CWVTE0055W: Initialize environment failed

CWVTE0056W: eciSetOutputDevice failed

CWVTE0057W: returning: NULL_ECI_HAND

300
CWVTE0058W: returning: NULL_ECI_HAND

CWVTE0059W: returning: false

CWVTE0060W: Reset environment and active voice failed - returning false

CWVTE0061W: returning: ECI_PARAMETERERROR

301
CWVTE0062W: Demonstration version - returning false

CWVTE0063W: SpeakText - eciNewEx failed - returning false

CWVTE0064W: SpeakText - eciAddText failed - returning false

CWVTE0065W: SpeakTextEx - eciNewEx failed - returning false

CWVTE0066W: Speaking error - returning false

302
CWVTE0067W: NULL eciHandle - returning false

CWVTE0068W: Invalid parameter - eciPhrasePrediction not supported returning -1

CWVTE0069W: Invalid parameter - eciNumDeviceBlocks not supported - returning -1

CWVTE0070W: setOutputToDevice failed - returning -1

303
CWVTE0071W: eciSizeDeviceBlocks not supported - returning -1

CWVTE0072W: eciSizePrerollDeviceBlocks not supported - returning -1

CWVTE0073W: eciNumVoiceParams not supported - returning -1

CWVTE0074W: Unable to allocate memory

Explanation: Unable to allocate memory. User response: If your Unix computer has adequate memory, increase the limit for physical memory if possible (see ulimit -m).
CWVTE0075W: Romanizer error

304
CWVTE0076W: createTextElement failed - returning false

CWVTE0077W: processManualQueue failed - returning false

CWVTE0078W: sendParameters failed - returning false

CWVTE0079W: calloc failed - returning false

CWVTE0080W: eciAddText2 failed - returning false

305
CWVTE0081W: NULL eciHandle or NULL text - returning false

CWVTE0082W: createIndexElement failed - returning false

CWVTE0083W: eciInsertIndex2 failed - returning false

CWVTE0084W: eciHandle is NULL

CWVTE0085W: File is empty - returning true

306
CWVTE0086W: NULL eciHandle or file open failed - returning false

CWVTE0087W: No file system support

CWVTE0088W: setOutputToPhonemeCallback or processManualQueue failed - returning false

CWVTE0089W: setOutputToSampleCallback failed - returning false

307
CWVTE0090W: ECIERROR_SYNTHESIS_ABORT - returning false

CWVTE0091W: NULL eciHandle or no callback - returning false

CWVTE0092W: Synthesizing - returning NULL handle

CWVTE0093W: returning: DictInvalidVolume

CWVTE0094W: returning: %s
308
CWVTE0095W: key or translation size 0 - returning DictNoEntry

CWVTE0096W: returning: DictAccessError

CWVTE0097W: size is 0
CWVTE0098W: returning: DictNoEntry

CWVTE0099W: NULL parameter

309
CWVTE0100W: returning: false

CWVTE0101W: returning: FilterAccessError

CWVTE0102W: Synthesizing - returning NULL_FILTER_HAND

CWVTE0103W: returning: ECI_PARAMETERERROR=%s

310
CWVTE0104W: Synthesizing - unable to register voice

CWVTE0105W: ECIERROR_ERROR
CWVTE0106W: ECIERROR_NORESOURCES
CWVTE0107W: PARAMETERERROR
CWVTE0108W: DEVICEERROR
311
CWVTE0109W: ECIERROR_AUDIODEVICEBUSY
CWVTE0110W: ECIERROR_ENGINEERROR
CWVTE0111W: ECIERROR_SYNTHESIZING
CWVTE0112W: Language or Romanizer not found

312
CWVTE0113W: sendChangedParameters or activeVoice failed returning false

CWVTE0114W: eciSynthesizeFile: NULL eciHandle or couldnt open file

CWVTE0150E: failed changing SPR type (esp2)

CWVTE0151E: failed changing SPR type (espr0 esp1)

CWVTE0152E: Your license has expired

313
CWVTE0153E: Could not find LUM license database file

CWVTE0154E: Invalid UTF8 char %s

CWVTE0155W: Cannot load Romanizer

CWVTE0156W: Cannot load Torrent engine

314
CWVTE0157W: setEngineSampleRate failed

CWVTE0158W: m_hCurrentEngine is NULL

CWVTE0159W: registerPhonemeBuffer failed

CWVTE0160W: m_uiCharsOnQueue <= 0

CWVTE0161W: This is an eval license of Text-to-Speech - you have %s days left

315
CWVTE0301E: Problem allocating bin_seq_binary structure

CWVTE0302E: Problem allocating binseq array for %s bytes

CWVTE0303E: Attempt to write beyond the number of bits available

CWVTE0311E: Problem allocating struct lpc_info

316
CWVTE0312E: Problem allocating autocorrelation array

CWVTE0313E: Problem allocating calculation array

CWVTE0314W: Warning k[%2$s] has value %3$s - forcing -1 < k < 1

CWVTE0315W: Warning E[%2$s] has value %3$s - forcing E[n] = 0

CWVTE0316E: Requested LP_order %2$s greater than initialization LP_order %3$s

317
CWVTE0321E: Error calling comp_f0_contour_ee_init

CWVTE0322E: Problem interpreting synthesis pitch

CWVTE0323E: Synthesis pitch out of range - now set to 100

CWVTE0324E: Problem interpreting pitch track filename

318
CWVTE0325E: The sentence total_leaf_duration_samples has not been computed

CWVTE0326E: comp_f0_contour initialization has not been successfully completed

CWVTE0327E: synthesis_mode & 2 set but no pitch_track_fn specified

CWVTE0328E: Problem opening pitch_track_fn %s

CWVTE0329E: Problem generating pitch track using ACs code

319
CWVTE0330E: You must compile with flag AARONF0 set to include ACs F0 generation code
CWVTE0331E: Problem generating pitch track using EEs code

CWVTE0332E: Problem generating pitch track using etis pitch

CWVTE0341E: No words specified

320
CWVTE0342E: No phones specified

CWVTE0343E: Problem opening ./word_file

CWVTE0344E: Problem allocating long

CWVTE0345E: Problem opening ./phrloc_file

CWVTE0346E: Problem opening ./accloc_file

321
CWVTE0347E: Problem opening ./aaron_temp.num_tags

CWVTE0348E: Problem opening ./aaron_temp.transcribe

CWVTE0349E: Unexpected EOF reading ./aaron_temp.num_tags

CWVTE0350E: Unexpected error reading ./aaron_temp.num_tags

CWVTE0351E: Problem reading ./aaron_temp.transcribe

322
CWVTE0352E: Problem determining number of syllables from ./aaron_temp.transcribe entry

CWVTE0353E: No accents types where expected

CWVTE0354E: Problem interpreting accent type

CWVTE0355E: Problem opening pitch_track_fn ./aaron_temp.final_pitch

323
CWVTE0356E: Problem reading pitch_track_fn ./aaron_temp.final_pitch

CWVTE0357E: Problem - trying to use eti pitch with non-eti phones

CWVTE0358E: Problem - ETI SPR did not contain a pitch contour

CWVTE0359E: Problem - ps->etiphrase->phone_durs is still NULL

CWVTE0361E: dur_leaf_id %s is out of the acceptable range

324
CWVTE0362E: Pnsp, voice, and broadclass list file (aux_dur_list_file) not specified
CWVTE0363E: Cannot open voicing map file %s

CWVTE0364E: Phone order in %s does NOT match phone order in PNSP file - re-order
CWVTE0365E: Cannot allocate for num_TAGS

325
CWVTE0366E: Phone number %s exceeds voicing table size

CWVTE0367E: Phone number %s exceeds broadclass table size

CWVTE0368W: Failed to allocate memory for broadclassmap array

Explanation: Unable to allocate memory. User response: If your Unix computer has adequate memory, increase the limit for physical memory if possible (see ulimit -m).
CWVTE0369E: DCSTree new failed for aux_dur_tree

CWVTE0370E: Cannot open aux leaf durs file %s

326
CWVTE0371E: Failed to allocate memory for auxdur

CWVTE0372E: Failed to allocate memory for auxdurwt

CWVTE0373E: dur_factor not specified in DDINF

CWVTE0374E: DCSTree new failed for dur

CWVTE0375E: DCSTree init failed for dur

327
CWVTE0376E: Duration medians not specified in the DDINF

CWVTE0377E: Problem opening input file %s

CWVTE0378E: Problem reading number of duration medians from disk

CWVTE0379E: Number of leaves in duration tree doesnt match number of medians in dur_medians file
328
CWVTE0380E: Problem allocating dur_medians array

CWVTE0381E: dur_phn_context_size not specified in DDINF

CWVTE0382E: Problem interpreting duration factor

CWVTE0383E: No phones specified in sentence structure

CWVTE0384E: Problem computing duration of phone %s

329
CWVTE0385E: An out of range dur_leaf_id occurred in this function

CWVTE0386E: Problem computing duration of leaf %s

CWVTE0387E: No leaves specified in sentence structure

CWVTE0388E: comp_leaf_durs initialization has not been successfully completed

330
CWVTE0389E: Problem in useETIdurs()

CWVTE0390E: Problem - trying to use eti durations with non-eti phones

CWVTE0391E: Problem descending state duration trees

CWVTE0392E: Problem descending phone duration trees

CWVTE0393E: Problem computing the duration of phone %s

331
CWVTE0394E: Problem with that input in quantizePhoneDurations aborting

CWVTE0395E: Problem computing the duration of leaf %s

CWVTE0396E: synthesis_mode & 4 set but no phones specified in sentence structure

CWVTE0401E: DCSTree new failed for en

332
CWVTE0402E: DCSTree init failed for en

CWVTE0403E: Energy medians not specified in the DDINF

CWVTE0404E: Problem reading number of energy medians from disk

CWVTE0405E: Number of leaves in energy tree doesnt match number of medians in en_medians file
CWVTE0406E: Problem allocating en_medians array

333
CWVTE0407E: en_phn_context_size not specified in DDINF

CWVTE0408E: An out of range en_leaf_id occurred in this function

CWVTE0409E: leaf_id %s is out of the acceptable range

CWVTE0410E: energy leaf name has no ( in it!?

334
CWVTE0411E: energy leaf name has no ) in it!?

CWVTE0412E: Problem interpreting energy leaf name %s

CWVTE0413E: en_leaf_id %s is out of the acceptable range

CWVTE0414E: Problem computing energy of leaf %s

CWVTE0415E: comp_leaf_ens initialization has not been successfully completed

335
CWVTE0416E: Problem descending energy trees

CWVTE0417E: Problem computing the energy of leaf %s

CWVTE0421E: vowel_list_file is not defined in the DDINF

CWVTE0422E: cannot open vowel file %s

336
CWVTE0423E: error reading vowel file

CWVTE0424E: Cannot allocate for ellen_vowels

CWVTE0425E: Invalid Vowel in the vowel file

CWVTE0426E: Cannot allocate for vowel file

CWVTE0427E: Cannot allocate for posmap array

337
CWVTE0428E: Cannot allocate for posmap array element

CWVTE0429E: Cannot open Part-Of-Speech file %s

CWVTE0430E: cannot allocate memory for testing_utt_tree.context

Explanation: Unable to allocate memory. User response: If your Unix computer has adequate memory, then if possible increase the limit for physical memory (see ulimit -m).
CWVTE0431E: DCSTree new failed for pitch_tree

CWVTE0432E: DCSTree init failed for pitch_tree

338
CWVTE0433E: leaf_pitches_deltas file should be specified in the ddinf

CWVTE0434E: cannot open pdfs file %s

CWVTE0435E: cannot allocate bestpitch

CWVTE0436E: cannot allocate deltapitch

CWVTE0437E: error reading pitch data file

339
CWVTE0438E: Failed to load vowel_list_file

CWVTE0439E: Failed load vowel_list_file

CWVTE0440E: cannot allocate accent_sequence

CWVTE0441E: cannot allocate this_pitch_contour

340
CWVTE0442E: cannot compute test features from eti

CWVTE0443E: cannot allocate ppx

CWVTE0444E: cannot allocate pitch elements

CWVTE0451E: allocating struct decom

CWVTE0452E: Problem allocating corr array

341
CWVTE0453E: Problem allocating offsets array

CWVTE0454E: Cant initialize LPC calculator

CWVTE0455E: i just got bigger than dec->frm_shift_samps

CWVTE0456E: LP_order %2$s exceeds the init value of %3$s

CWVTE0461E: There is no RawInput stream

342
CWVTE0462E: There is no inp stream

CWVTE0463E: Problem allocating struct syn_phone array

CWVTE0464E: There is no word stream

CWVTE0465E: There is no phone stream

343
CWVTE0466E: There is no syllable stream

CWVTE0467E: There is no Ms stream

CWVTE0468E: There is no F0 stream

CWVTE0469E: There is no offset stream

CWVTE0470E: There is no inton_phr stream

344
CWVTE0471E: There is no control stream

CWVTE0472E: Failed to allocate %s etiSprWord structure

CWVTE0473E: Failed to allocate %s etiSprSyllable structure

CWVTE0474E: Failed to allocate %s etiSprPhone structure

CWVTE0475E: Failed to allocate %s etiF0 structure

CWVTE0476E: Phone stream has more phones (%2$s) than we calculated earlier (%3$s)
345
CWVTE0477W: There are no words in the wordstream

CWVTE0478W: There are no phones in the phonestream

CWVTE0479W: There are no words in the wordstream

CWVTE0480E: Problem allocating struct etiSprPhrase array

CWVTE0481E: Failed to allocate %s etiSprWord structure

346
CWVTE0482E: Failed to allocate %s etiSprSyllable structure

CWVTE0483E: Failed to allocate %s etiSprPhone structure

CWVTE0484E: Failed to allocate %s etiF0 structure

CWVTE0491E: Problem starting ETI engine

CWVTE0492E: Problem with eciSetParam eciLanguageDialect

347
CWVTE0493E: Could not load language-specific main dictionary into ETI

CWVTE0494E: Could not load language-specific root dictionary into ETI

CWVTE0495E: Could not load language-specific abbreviation dictionary into ETI

CWVTE0496E: Could not set ETI to use language-specific exeptions dictionary(s)

348
CWVTE0497E: Problem allocating EtiFE class

CWVTE0498E: Calling genBaseform when not initialized

CWVTE0499E: No text supplied to genBaseform()

CWVTE0500E: Calling genBaseform() with no non-white-space text

CWVTE0501E: Problem allocating eti_text array

349
CWVTE0502E: Problem calling eciAddText()

CWVTE0503E: Problem allocating retbuf array

CWVTE0511E: Cannot allocate work array

CWVTE0512E: Cannot allocate ip array

350
CWVTE0513E: Problem allocating data array

CWVTE0514E: Sorry, you have to call FD_sigproc_init() at initialization time

CWVTE0515E: Problem allocating mp array

CWVTE0516E: Problem allocating fft cwin array

CWVTE0517E: Problem allocating fft cmp array

351
CWVTE0518E: Problem allocating har array

CWVTE0521E: Problem allocating huf table

CWVTE0522E: Problem allocating huf elems

CWVTE0523E: problem allocating a hash element

352
CWVTE0524E: Problem opening output file %s

CWVTE0525E: Problem allocating huff table for %s

CWVTE0526E: Problem allocating code char array

CWVTE0527E: problem writing to binary sequence

CWVTE0531W: at leave %s there are is zero (prevactive = 0) active nodes in the beam
353
CWVTE0532E: No solution for %s

CWVTE0533E: One of num_U_marks or num_V_marks is 65535

CWVTE0534E: U marks required but none in occurrence

CWVTE0535E: V marks required but none in occurrence

354

CWVTE0537E: base_synth initialization has not been successfully completed

CWVTE0538E: leaves2occs initialization has not been successfully completed

CWVTE0539E: synthesis_mode says use continuity costs - but vectors are not loaded
CWVTE0540E: There are unspecified durations in the leaf level of the sentence structure
355
CWVTE0541E: synthesis_mode says do pitch mark based duration costing but the pitch mark files were never processed
CWVTE0542E: Problem allocating pitch_elem array

CWVTE0543E: Problem calling occs2speech_reset()

CWVTE0544E: Problem allocating vit_info structure

356
CWVTE0545E: Problem allocating vit_elem array

CWVTE0546E: Attempt to synthesise leaf %s which has no occs

CWVTE0547E: Problem allocating struct cand_occ array

CWVTE0548E: Problem allocating tracer array

CWVTE0549E: Problem in do_stuff() for leave %s

357
CWVTE0550E: Problem calling occs2speech_process

CWVTE0551E: Problem calling occs2speech_cleanup()

CWVTE0552E: Problem opening input file %s

CWVTE0553E: Problem reading input file %s

CWVTE0554E: Problem allocating vecs->coef pointer array

358
CWVTE0555E: Number of vectors loaded from %2$s is %3$s which does not match number expected by core occurrence files: %4$s
CWVTE0556E: Problem reallocing vecs->coef pointer array

CWVTE0557E: Vector %2$s has dimension %3$s which is different from the first vector with dimension %4$s
CWVTE0558E: Problem allocating vecs->coef Vcoef array

359
CWVTE0559E: Problem reading num join leafs from file %s

CWVTE0560E: Problem reading join vec_dim from file %s

CWVTE0561E: Dimension in join file %s doesnt agree with that in vectors file
CWVTE0562E: Problem allocating join pointer array

CWVTE0563E: Problem allocating join array

360
CWVTE0564E: Problem reading join data

CWVTE0565E: join_matrix not specified in DDINF

CWVTE0566E: join_means not specified in DDINF

CWVTE0567E: join_stdevs not specified in DDINF

361
CWVTE0568E: Problem allocating the join structure

CWVTE0569E: Problem reading num_phones

CWVTE0570E: num_phones in join matrix %2$s differs from that in phonegraphs %3$s
CWVTE0571E: Problem reading num_internals

CWVTE0572E: num_internals in join matrix %2$s is different from that implied by phonegraphs %3$s
362
CWVTE0573E: Problem allocating join matrix

CWVTE0574E: Problem reading join matrix data

CWVTE0575E: Problem reading join means

CWVTE0576E: Problem reading join stdevs

363
CWVTE0577E: integer_f0_value not specified in DDINF

CWVTE0578E: Problem reading in vector list file %s

CWVTE0579E: Problem loading join stats

CWVTE0580E: Command longer than 1023 characters - aborting

CWVTE0581E: No command specified

364
CWVTE0582E: cannot open usage stats file %s for saving

CWVTE0583E: error saving the usage stats file

CWVTE0584E: no file name specified

CWVTE0585E: cannot open usage stats file %s for saving

CWVTE0586E: cannot open usage stats file %s for loading

365
CWVTE0587E: error reading usage stats file

CWVTE0588E: usage stats file does not match the loaded dataset
CWVTE0589E: Usage: *l2op <param_name> <param_value> <CR> or *l2op ? <CR>

CWVTE0590E: Usage: l2op <param_name> <param_value>

366
CWVTE0591W: No vector_list_fn specified, setting to NULL

CWVTE0592W: Beware, this compilation will read vectors as %s byte values

CWVTE0593W: this command loads usage stats incremently; * zero_stat should be used prior to first use
CWVTE0600E: Problem allocating syn_leaf array

CWVTE0610E: phfn file contains less than 5 phones - aborting

367
CWVTE0621E: cannot open multiple ebuf file %s

CWVTE0622E: Problem allocating ebuf array

CWVTE0623E: Problem with sentence %s in synthesise()

CWVTE0624E: language not specified in DDINF

368
CWVTE0625E: Problem allocating phone_durs array

CWVTE0626E: num != ps->num_phones

CWVTE0627E: Problem allocating lpfehash_elem structure

CWVTE0628E: Maximum ETI phone length is 9 characters in file %s

CWVTE0629E: map_fn not specified in DDINF

369
CWVTE0630E: Problem opening map file %s

CWVTE0631E: Problem reading file %s

CWVTE0632E: Problem talking to ETI dll

CWVTE0633E: Problem calling map_eti2ibm_init()

CWVTE0634E: Stress is expected to be 0, 1 or 2 - it is %s

370
CWVTE0635E: eti phone %s not found in hash table

CWVTE0636E: No phones specified in ETI SPR

CWVTE0637E: Problem allocating struct syn_phone array

CWVTE0638E: Problem in map_eti2ibm()

371
CWVTE0639E: The ETI buffer is ill defined - it does not have three null chars
CWVTE0640E: SPR specified is empty

CWVTE0641E: Problem allocating sentence array

CWVTE0642E: Problem allocating spr array

CWVTE0643E: Problem allocating etiSprPhrase structure

372
CWVTE0644E: Problem parsing the ETI SPR %s

CWVTE0645E: Error in processing CN eti phrase

CWVTE0646E: Error in processing CT eti phrase

CWVTE0647E: Error in processing Korean eti phrase

373
CWVTE0648E: Problem in mapping ETI phones to IBM phones

CWVTE0649E: Problem computing durations from SPR structure

CWVTE0650E: Problem calling process ebuf

CWVTE0651E: Problem generating an ebuffer from the ETI dll

CWVTE0652E: Direct Access to ETI is not allowed in Torrent

374

CWVTE0662E: occs2speech initialization has not been successfully completed

CWVTE0663E: Problem allocating %s shorts to hold synthetic speech

CWVTE0664E: RC=%2$s Line=%3$s

375
CWVTE0665E: Problem allocating memory for segment cue-marks

CWVTE0666E: cell %s does not exist in stem file

CWVTE0667E: trying to synthesis from feature file %s which doesnt exist

CWVTE0668E: problem querying frame number - RC=%2$s line=%3$s

CWVTE0669E: RC=%2$s Line=%3$s GAIN_FACTOR

376
CWVTE0670E: Problem forgetting history - RC=%2$s line=%3$s

CWVTE0671E: Problem accessing data for cell %s

CWVTE0672E: Problem writing wave file with segment marks %s

CWVTE0673E: Problem allocating wav_info structure

377
CWVTE0674W: Synthesized speech reached ABS_TIME_LIMIT

CWVTE0675W: Ospeech_loc = %2$s, requested leaf dur = %3$s, ABS_TIME_LIMIT = %4$s

CWVTE0676W: Synthesized speech reached ABS_TIME_LIMIT while generating silence

CWVTE0681E: ext_pron_dict not specified in DDINF

CWVTE0682E: occ_file_dir not specified in DDINF

378
CWVTE0683E: Problem in xwal2ps_init()

CWVTE0684E: Problem opening occurrence file %s

CWVTE0685E: occ file %2$s appears to be longer than the xwal file %3$s
CWVTE0686E: alloarc name has no ( in it!?

379
CWVTE0687E: occ file fenemes in %2$s have got out of step with xwal fenemes in %3$s
CWVTE0688E: alloarc name has no ) in it!?

CWVTE0689E: Problem interpreting name %s

CWVTE0690E: Problem in init()

CWVTE0691E: Problem loading xwal file %s

380
CWVTE0692E: Problem in add_occ_file_info()

CWVTE0701E: chunk %s damaged at beginning

CWVTE0702E: chunk %s damaged at end

CWVTE0703E: Problem allocating chunks pointer array

381
CWVTE0704E: Problem allocating chunk[%s]

CWVTE0705E: Problem reallocating chunk[%s]

CWVTE0706E: problem, attempt to allocate space of size 0

CWVTE0707E: problem flushing chunk %s

CWVTE0708E: Problem reallocating chunks pointer array

382
CWVTE0709E: chunk_pos %2$s is greater than this_chunks_size %3$s

CWVTE0710E: Problem, realloc has moved chunk %s

CWVTE0721E: Problem huff decoding LP_order

CWVTE0722E: Problem huff decoding LP coeff %s

CWVTE0723E: Problem huff decoding free_flow

383
CWVTE0724E: Problem huff decoding offset %s

CWVTE0725E: Problem huff decoding quant

CWVTE0726E: Problem reading from binary sequence

CWVTE0727E: Problem in decompress() - aborting

384

CWVTE0729E: Problem allocating o2s_process structure

CWVTE0730E: Problem allocating this_window array

CWVTE0731E: occs2speech initialization has not been successfully completed


385
CWVTE0733E: There are unspecified durations in the leaf level of the sentence structure
CWVTE0734E: There are unspecified occurrences in the leaf level of the sentence structure
CWVTE0735E: Problem reallocating %s shorts to hold synthetic speech

CWVTE0736E: Attempt to use occurrence in synthesis for which no valid synthesis information (pitch-marks) exists.
386
CWVTE0737E: Problem allocating syn_mark structure

CWVTE0738E: Occurrence duration of leaf %2$s occurrence_id %3$s is <= 0

CWVTE0739E: psm->occ_mark is still -1 after finding the closest mark

CWVTE0740E: Problem determining pitch at time (samples) %s

CWVTE0741E: Problem loading waveform file %s

387
CWVTE0742E: preload is 1, and yet waveforms[%s] is NULL

CWVTE0743E: preload is 2, but no Cwaveform loaded for cell %s

CWVTE0744E: Problem allocating short decompression array

CWVTE0745E: Call to frequency domain signal processing code failed

388
CWVTE0746E: Problem calling signal processing streaming function stream_speech()

CWVTE0747E: Problem converting short data into mulaw data

CWVTE0748E: Problem opening mark file %s

CWVTE0749E: Problem allocating file_marks structure array

CWVTE0750E: Problem accessing pMarkInfoLinkArray memory from CMM

389
CWVTE0751E: Problem accessing pPitchMarkLinkArray memory from CMM

CWVTE0752E: num_pitch_marks is too large

CWVTE0753E: Problem allocating mark_info structure

CWVTE0754E: Problem allocating pitch_mark structure array

390
CWVTE0755E: T mark as the first mark in mip array

CWVTE0756E: Two T marks in a row

CWVTE0757E: The mip array ends in a T mark

CWVTE0758E: wav_file_dir not specified in DDINF

CWVTE0759E: wav_file_ext not specified in DDINF

391
CWVTE0760E: Preload not specified in DDINF

CWVTE0761E: Cwav_file_dir not specified in DDINF

CWVTE0762E: Cwav_file_ext not specified in DDINF

CWVTE0763E: frm_shift_samples not specified in DDINF

CWVTE0764E: Problem allocating check_data array

392
CWVTE0765E: tree1_fn not specified in DDINF



CWVTE0768E: S not specified in DDINF

393
CWVTE0769E: max_LP not specified in DDINF

CWVTE0770E: quantise not specified in DDINF

CWVTE0771E: Problem allocating decomp_buf array

CWVTE0772E: problem loading huffman tree %s

CWVTE0773E: problem in decompress_init() - aborting

394
CWVTE0774E: mark_file_dir not specified in DDINF

CWVTE0775E: psla_time_samples not specified in DDINF

CWVTE0776E: volume not specified in DDINF

CWVTE0777E: Problem loading marks file %s

CWVTE0778E: Problem allocating bin structure

395
CWVTE0779E: Problem reading in speechfile %s

CWVTE0780E: Can not allocate han array

CWVTE0781E: Problem calling FD_sigproc_init()

CWVTE0782E: Problem calling post_filter_init()

396
CWVTE0783E: Problem opening specified list_of_bad_occ_indices file, %s

CWVTE0784E: It appears that leaf %s has no useable occs

CWVTE0785E: post_filter not specified in DDINF

CWVTE0786E: Problem opening post_filter file %s

CWVTE0787E: Problem allocating post_filter structure

397
CWVTE0788E: Problem reading lnum from post_filter file %s

CWVTE0789E: Problem allocating F->num array

CWVTE0790E: Problem reading F->num from post_filter file %s

CWVTE0791E: Problem reading lden from post_filter file %s

398
CWVTE0792E: Problem allocating F->den array

CWVTE0793E: Problem reading F->den from post_filter file %s

CWVTE0794E: Problem allocating F->buf array

CWVTE0795E: Failed to allocate %s bytes for agc_parm structure

CWVTE0796W: No pitch marks in file %2$s for leaf %3$s at time %4$s to time %5$s
399
CWVTE0800E: getObject failed to create TorrentWrap object

CWVTE0810E: Problem allocating double array of %s samples

CWVTE0820E: No extension stems file specified - createcustomevoice *command failed

CWVTE0821E: Problem opening stem file %s

400
CWVTE0822E: Problem allocating extstem char array index %s

CWVTE0823E: No extension vectorlist file specified createcustomevoice *command failed

CWVTE0824E: Attempt to use samp2time before initializing sample_rate

CWVTE0825E: Attempt to use time2samp before initializing sample_rate

CWVTE0826E: Trying to look up pitches before specifying a syn_pitch structure

401
CWVTE0827E: Trying to look up pitches before specifying pitch_elems

CWVTE0828E: Cannot find pitch for time %s in the pitch structure

CWVTE0829E: There are two pitch_elems with the same time value
CWVTE0830E: Just called init_prefix() with a NULL

402
CWVTE0831E: Problem allocating new filename char array

CWVTE0840E: Cmm::initialize failed

CWVTE0841E: CMM failed to allocate %s bytes of shared memory

Explanation: Unable to allocate the requested amount of shared memory. User response: If loading several voices, reduce the number and try again. On Unix check the owner and permission attributes of shared memory (ipcs -m).
CWVTE0842E: MaxChunk size is %2$s, but requested memory size by the dataset is %3$s, which is an invalid condition
CWVTE0843E: Failed to allocate memory of size %s

403
CWVTE0844E: CMM Failed to allocate memory of size %s

CWVTE0845E: Trying to request more memory than allocated, currentpos = %2$s, bytes requested = %3$s, TotalPSMemoryAllocated = %4$s.
CWVTE0846E: Memory request of %2$s bytes exceeds maximum allocated (%3$s)

CWVTE0847E: Another process is still loading the dataset %s

CWVTE0848E: Failed to create/open %s file

404
CWVTE0861E: problem adding %s.occ file to the splice_reduction hashtable

CWVTE0871E: cmm::releaseNow() failed for 0x%s after addCache failed to add it into cache array
CWVTE0872E: cmm returned %2$s while creating memory for cache item %3$s - the return code should be %4$s
CWVTE0873E: cmm::releaseNow() failed for 0x%s after using cached item

405
CWVTE0874E: cmm::releaseNow() failed for 0x%s while making space for a new cache item (using # of bytes method)
CWVTE0875E: Did not find the named (%2$s) memory while removing a item# %3$s
CWVTE0876E: failed to write cachefile correctly - attempted to write %2$s items but only wrote %3$s items - clearing the cache file
CWVTE0881E: ph_cntx[pos] is incorrect because %2$s (pos) + %3$s (offset) is a negative value
406
CWVTE0891E: Invalid configuration file name: %s

CWVTE0892E: Failed to allocate a buffer %s for Param

CWVTE0911E: Failed while loading SynthInfo information from OFFLImage file

CWVTE0912E: Failed to read ReadOnlyMemory information from OFFLImage file

CWVTE0913E: Fastload image version %2$s does not match with the fastload build engine version %3$s
407
CWVTE0914E: Unrecognised command (%s)

CWVTE0915E: syn_sentence structure was not created from reading Delta

CWVTE1501E: unable to open file: %2$s errno = %3$s

CWVTE1502E: client: Unable to create initialization mutex semaphore

408
CWVTE1503E: client: unable to start memory manager: %2$s rc = %3$s

CWVTE1504E: client: unable wait for memory manager

CWVTE1505E: Access to startup semaphore failed

CWVTE1506E: Post failed

CWVTE1507E: Wait failed for semaphore - rc = %s

409
CWVTE1508E: Release failed for semaphore - rc = %s

CWVTE1509W: psubAlloc = %s was not allocated

CWVTE1510E: CreateProcess failed - error = %s

CWVTE1511E: OpenProcess failed for pid %2$s - error = %3$s

410
CWVTE1512E: WaitForSingleObject failed - error = %s

CWVTE1513E: InitializeSecurityDescriptor failed - error = %s

CWVTE1514E: SetSecurityDescriptorDacl failed - error = %s

CWVTE1515E: CloseHandle failed - handle = %2$s - error = %3$s

CWVTE1516E: Server segment not found in client access list pserverSharedMemorySegment = %s

411
CWVTE1517E: Client segment not found in client access list pclientSharedMemorySegment = %s

CWVTE1518W: attached to a previously attached segment

CWVTE1519E: OpenFileMapping failed - error = %s

CWVTE1520E: MapViewOfFile failed - error = %s

412
CWVTE1521E: DuplicateHandle failed - error = %s

CWVTE1522E: GetExitCodeProcess failed - handle = %2$s error = %3$s

CWVTE1523E: CreateFileMapping failed - error = %s

CWVTE1524E: getSharedMemory failed

CWVTE1525E: getNamedSharedMemory failed

413
CWVTE1526E: shmctl failed - errno = %s

CWVTE1527E: shmat failed - unable to attach shared memory - errno = %2$s : %3$s
CWVTE1528E: inaccessible memory has key: 0x%2$s created by user: %3$s group: %4$s
CWVTE1529E: ftok failed for file name: %2$s and id = %3$s

414
CWVTE1530E: shmget failed - memory not found - errno = %s

CWVTE1531E: shmget failed - invalid amount of memory requested: %s

CWVTE1532E: shmget failed - errno = %s

CWVTE1533E: memory with key: 0x%2$s created by user: %3$s group: %4$s - attempted access by user: %5$s group: %6$s.
CWVTE1534E: fork failed - errno = %s

415
CWVTE1535E: execlp failed - errno = %s

CWVTE1536E: ibmcmm was not started by this instance of cmmcmd

CWVTE1537E: ibmcmm terminated by %s signal

CWVTE1538E: ibmcmm terminated with return code: %s

416
CWVTE1600E: createSemaphore failed - rc = %s

Explanation: Unable to create semaphore - exiting. User response: See previously logged related messages.
CWVTE1601E: setSemaphore failed - rc = %s

Explanation: Unable to set value of semaphore. User response: See previously logged related messages.
CWVTE1602E: resetSemaphore failed - rc = %s

Explanation: Unable to reset value of semaphore. User response: See previously logged related messages.
CWVTE1603E: waitForSemaphore failed - rc = %s

Explanation: Wait for semaphore failed. User response: See previously logged related messages.
CWVTE1604E: getSemaphore failed - rc = %s

Explanation: Unable to allocate or access private semaphore. User response: See previously logged related messages.
CWVTE1605E: unexpected semaphore error: rc = %s

Explanation: Unexpected error encountered while recovering from previous error. User response: See previously logged related messages.
CWVTE1606E: semop failed - semid = %2$s errno = %3$s

Explanation: semop call failed. User response: See value of errno in errno.h to determine reason.
CWVTE1607E: sigaction failed - errno = %s

417
CWVTE1608E: setitimer failed - errno = %s

CWVTE1609E: semctl failed - errno = %2$s, sema4Id = %3$s

CWVTE1610E: semaphore: owner ID = %2$s, owner group = %3$s, permission bits = %4$s, creators ID = %5$s
Explanation: semop failed. The semaphore owner, access bits, and creators ID are listed. User response: Verify that process attempting to access the semaphore has permission to do so.
CWVTE1611E: processs: real user ID = %2$s, effective user ID = %3$s

Explanation: semop failed. The current processs real and effective IDs are listed. User response: This information helps to determine if the process has permission to access a semaphore.
CWVTE1612E: ftok failed - name = %2$s id = %3$s errno = %4$s

Explanation: ftok system call failed. name is the name of the target file name, and errno is the reason for the failure. User response: The value of errno fives the reason the system call failed for the file. For example, 2 means file not found and 13 means permission denied. See errno.h for more information.
CWVTE1620E: OpenProcess failed - error = %s

418
CWVTE1621E: DuplicateHandle failed - error = %s

CWVTE1622E: SetEvent failed - error = %s

CWVTE1623E: ResetEvent failed - error = %s

CWVTE1624E: WaitForSingleObject failed - error = %s

419
CWVTE1625W: CloseHandle failed - error code = %s

CWVTE1626E: CreateEvent failed - error = %s

CWVTE1627E: ReleaseMutex failed - error code = %s

CWVTE1628E: InitializeSecurityDescriptor failed - error = %s

CWVTE1629E: SetSecurityDescriptorDacl failed - error = %s

420
CWVTE1630E: CreateMutex failed - error code = %s

CWVTE1640E: Error allocating memory

CWVTE1642W: returning: ECI_PARAMETERERROR=%s

CWVTE1645W: Could not convert character code %s to its corresponding UTF8 char
CWVTE1646W: Ill formed UTF-8 character:`%s` - processing to this point

421
CWVTE1650E: Error creating ssml processor instance

CWVTE1651E: Error creating parser instance

CWVTE1652E: Unable to set initial filter environment, eci environment not available
CWVTE1661E: Unable to allocate memory of size %s

CWVTE1662E: Error Allocating memory

422
CWVTE1663W: IPA phoneme alphabet not supported for current active language %s
CWVTE1664E: Unknown or unsupported value for <phoneme alphabet=%s>

CWVTE1665W: unsupported level: %s

CWVTE1666W: Unknown or unexpect <say-as> interpret-as vxml:date format %s, tag will be ignored
CWVTE1667W: Unknown or unsupported xml:lang value of %s

423
CWVTE1668W: Unknown or unavailable xml:lang value of %s, block ignored

CWVTE1669W: Error converting back the input Phoneme

CWVTE1670W: IPA input %s cannot be converted to IBM SPR

CWVTE1671W: Error converting the returned string from SPR mapping

424
CWVTE1672E: Error, requested <%2$s> with %3$s=%4$s not initialized

Explanation: Configure WebSphere Voice Server to initialize the voice User response: Verify that the WebSphere Voice Server provides file has your voice in the initialization list
CWVTE1673E: Error, requested <%2$s> with %3$s=%4$s not available

Explanation: The requested value is not available User response: You may be asking for an invalid language or one that is not currently installed
CWVTE1676E: <prosody> tag with attribute duration not supported

CWVTE1677E: <prosody> tag with attribute contour not supported

CWVTE1678W: Unable to obtain previous prosody pitch value

CWVTE1679W: Unable to obtain previous prosody range value

425
CWVTE1680W: Unable to obtain previous prosody rate value

CWVTE1681W: Unknown or unsupported gender=%s, attribute will be ignored

CWVTE1682E: Error, closing voice tag may be unmatched

CWVTE1683W: Unknown or unsupported say-as attribute %s, tag ignored

426
CWVTE1684W: No format given for <say-as> interpret-as=number, tag ignored

CWVTE1685E: Element <%s> cannot occur in current content

Explanation: The input text generated an error. The SSML might not be compliant. User response: Verify the SSML input is compliant. Visit http://www.w3.org/TR/ speech-synthesis/ for more information on SSML.
CWVTE1686W: Error getting the languages installed

CWVTE1700E: unable to malloc %s bytes

CWVTE1701E: realloc failed for %s bytes

CWVTE1750E: cannot insert token object between the first & last two syncmarks: stream = %s
427
CWVTE1760E: Project failed for input %s

CWVTE1770W: invalid phone found

CWVTE1771E: malloc failed for %s bytes

CWVTE1780E: Failed to allocate memory for listenerList

CWVTE1781E: Failed to reallocate memory for listenerList

428
CWVTE1782E: Failed to allocate CEnvEventListener

CWVTE1783W: No active Torrent library

CWVTE1784W: No active Torrent instance

CWVTE1785E: Failed to load %s dataset

CWVTE1786E: Either Torrent library loading failed or could not create a TorrentWrap object
429
CWVTE1787E: getObject failed to create TorrentWrap

CWVTE1788E: getObject for Torrent library failed

CWVTE1789E: Failed to load %s library

CWVTE1790W: Did not find library path for Torrent version %s

430
CWVTE1791E: Failed to allocate CEnvEventListener

CWVTE1801E: Out of memory

CWVTE1802E: device not output and update

CWVTE1803W: device ID greater than 15 - deviceID = %s

CWVTE1804W: device ID < 0 - deviceID = %s

431
CWVTE1805E: ioctl: AUDIO_INIT failed - errno = %s

CWVTE1806W: start failed - errno = %s

CWVTE1807W: audio device is hung up with %s bytes remaining restarting

CWVTE1808E: (timediff > uPlaybackLeft) - timediff = %2$s uPlaybackLeft = %3$s

CWVTE1809E: ioctl: AUDIO_CONTROL failed - errno = %s

432
CWVTE1810E: ioctl failed - unable to stop device - errno = %s

CWVTE1811W: close failed for device: %2$s - errno = %3$s

CWVTE1812E: ioctl AUDIO_BUFFER failed - errno = %s

CWVTE1813E: write failed - errno = %s

433
CWVTE1821E: ioctl error executing SNDCTL_DSP_SETFRAGMENT

CWVTE1822E: Error setting number of channels

CWVTE1823E: ioctl failed SNDCTL_DSP_RESET - errno = %s

CWVTE1824E: ioctl error executing SNDCTL_DSP_GETODELAY

CWVTE1825E: ioctl error executing SNDCTL_DSP_GETOPTR

434
CWVTE1826E: Error releasing the audio device

CWVTE1827E: Error claiming an audio device

CWVTE1828E: Error ioctl SNDCTL_DSP_SPEED

CWVTE1829E: ioctl SNDCTL_DSP_SPEED could not set rate to %2$s, card may be locked at %3$s
435
CWVTE1830E: Error ioctl SNDCTL_DSP_STEREO

CWVTE1831E: Error setting number to %2$s channels, card set to do only %3$s
CWVTE1832E: Error ioctl SNDCTL_DSP_CHANNELS

CWVTE1833E: Sound card does not support mono, set to %s channels only

436
CWVTE1911W: Parser or input text is invalid

CWVTE1912W: Nodelist is not initialized

CWVTE1913W: index requested is out of range

CWVTE1914E: Error allocating memory in constructor

CWVTE1921E: Error allocating %s

CWVTE1922W: Invalid UTF8 sent in to parser

437
CWVTE1923W: Error found inside the attribute list of a tag

CWVTE1924E: Error while attempting to allocate memory by parser, returning processed text
CWVTE1925W: Mismatched tag found by parser, synthesizing only till illegal tag
CWVTE1926W: Unrecognized tag sent by parser: %s


Explanation: Unable to allocate memory.
438
User response: If your Unix computer has adequate memory, then if possible increase the limit for physical memory (see ulimit -m).
CWVTE2002W: Error deleting previous ECI dictionary

CWVTE2003W: Error deleting previous ECI handle

CWVTE2004W: Missing entry or counter is incorrect

CWVTE2005W: Error deleting incomplete entries

CWVTE2011W: eci handles are invalid: returning false

Explanation: Language may not be install User response: Make sure the language requested in installed
439
CWVTE2012W: Invalid parameter

Explanation: Invalid parmater, lexicon version 1.0 is supported User response: Check Lexicon files for 1.0 compliance
CWVTE2013W: Unknown or undefined lexicon tag

Explanation: Unrecognized tag in Lexicon User response: Make sure Lexicon is version 1.0 compliant
CWVTE2014W: Lexicon version=%s may not be supported

Explanation: Only version 1.0 is supported User response: Write your Lexicon using version 1.0
CWVTE2015W: Error adding LEXICON entry

CWVTE2016W: Error adding LEXEME entry

CWVTE2017W: Error adding SOUNDSLIKE entry

440
CWVTE2018W: Unknown or unrecognized tag <%s>

Explanation: Unrecognized tag in lexicon. User response: Make sure Lexicon is version 1.0 compliant.
CWVTE2019W: Unable to retrieve LEXICON from stack

CWVTE2020W: Possible missing tags missing in <lexeme>

Explanation: Make sure you have a valid lexicon file. User response: Make certain all the tags in your lexicon have a matching closing tag.
CWVTE2021W: Unable to retrieve PHONEME from stack

CWVTE2022W: Unable to retrieve SPELLING from stack

CWVTE2023W: Unable to retrieve SOUNDSLIKE from stack

441
CWVTE2024W: unknown closing tag</%s>

Explanation: Only version 1.0 tags are allowed User response: Make certain Lexicon 1.0 tag names are in lexicon file
CWVTE2025W: Unknown or unexpected type:%s

CWVTE2026W: Syntax Error in lexicon file

CWVTE2031E: Maximum word length exceeded %s

CWVTE2032E: error opening file %s

442
CWVTE2041W: Cannot process load-lexicon message - TTS media control layer may not be initialized.
Explanation: You must call TTS Media Control Layer with an initialize message first User response: Initialize failed or Initialize message not received
CWVTE2042W: An error occurred evaluating lexicon - URI->%s is not valid

CWVTE2043E: msg:invalid speak request, missing mcport data

Explanation: Required parameter mcport is not specified. User response: Specify a valid value for the mcport parameter.
CWVTE2044E: Failed on connecting to media converter

CWVTE2045W: invalid message: %s

443
CWVTE2046E: Error on creating thread for speak

CWVTE2047W: ECIINI path not specified - using %s

CWVTE2048E: Specify PATH to eci.ini file

CWVTE2049E: unable to update TTS configuration file

CWVTE2050E: msg:invalid speak request, missing kill-on-barge-in param

Explanation: Required parameter kill-on-barge-in missing. User response: Specify a valid kill-on-barge-in parameter.
444
CWVTE2051E: msg:invalid speak request, invalid kill-on-barge-in %s

Explanation: The specified kill-on-barge-in parameter is invalid. User response: Specify a valid kill-on-barge-in parameter.
CWVTE2052E: msg:Invalid speak request, missing audio-format param

Explanation: Required parameter audio-format missing. User response: Specify a valid value for the audio-format parameter.
CWVTE2053E: msg:invalid speak request, unknown audio-format %s

Explanation: The specified audio-format is invalid. User response: Specify a valid audio-format.
CWVTE2054E: TTS engine failed on initialize

Explanation: The TTS Engine was unable to initialize using the parameters specified. User response: Correct the invalid parameter value(s) and try the initialize request again.
CWVTE2055E: Failed retrieving URI: %s

CWVTE2056E: Tag <%2$s> is missing required attribute->%3$s

Explanation: SSML Tag missing a required element. User response: SSML specification 1.0 requires the missing atribute be present, modify your SSML input.
CWVTE2057E: Tag <%2$s> has invalid attribute %3$s=%4$s

Explanation: SSML Tag has invalid value for attribute. User response: The value being given is not valid for the attribute, modify SSML input.
445
CWVTE2058W: Cannot release shut down memory - setting server time-out

CWVTE2059W: Cannot query shut down memory - setting server time-out

CWVTE2060W: Cmm::query returned bad status for shut down memory

CWVTE2061E: Error on starting WSA

CWVTE2062E: Failed on getting host(TTS EJB) address, using IP=%2$s, errno=%3$s(%4$s)

446
CWVTE2063E: Error on creating socket for EJB messages errno=%2$s(%3$s)

CWVTE2064E: Failed on connecting to host EJB, using IP=%2$s, port=%3$s, errno=%4$s(%5$s)

CWVTE2065E: Failed on getting host MC address - using host MC IP=%2$s, state=%3$s, errno=%4$s(%5$s)
CWVTE2066E: Error on creating socket for MC (audio) - state=%2$s, errno=%3$s(%4$s)

447
CWVTE2067E: Failed on connecting to host MC (audio) - using audio port=%2$s, state=%3$s, errno=%4$s(%5$s)
CWVTE2068E: Error on creating socket for MC (message) state=%2$s, errno=%3$s(%4$s)

CWVTE2069E: Failed on connecting to host MC - using message port=%2$s, state=%3$s, errno=%4$s(%5$s)

CWVTE2070E: Error on creating thread for listening MC message state=%2$s, errno=%3$s(%4$s)

448
CWVTE2071E: Failed on getting host(RR) address, using IP=%2$s, errno=%3$s(%4$s)

CWVTE2072E: Failed on getting HTTP server

CWVTE2073E: Error on creating socket for IBM HTTP Server state=%2$s, errno=%3$s(%4$s)
Explanation: The SynthEngine can not create the connection to IBM HTTP Server. User response: Ensure the IBM HTTP Server is started.
CWVTE2074E: Failed on connecting to host IBM HTTP Server - using port=%2$s, state=%3$s, errno=%4$s(%5$s)
Explanation: The SynthEngine can not connect to the IBM HTTP Server. User response: Ensure the IBM HTTP Server is started.
CWVTE2075E: Unable to send a message to MC using port=%2$s(MC_msg_peer port=%3$s), state=%4$s, errno=%5$s(%6$s), msg=%7$s
449
CWVTE2076E: Unable to send audio to MC using audio port=%2$s(MC_audio_peer port=%3$s), state=%4$s, size of audio=%5$s, errno=%6$s(%7$s)
Explanation: The Media Converter might not be running on the expected port. User response: Verify that WebSphere Voice Server started successfully and check the error logs to make sure that the Media Converter component started without errors.
CWVTE2077E: Failed on sending message to EJB - state=%2$s, errno=%3$s(%4$s), msg=%5$s

CWVTE2078E: Failed on sending message to IBM HTTP Server state=%2$s errno=%3$s(%4$s) msg=%5$s
CWVTE2079W: Connection between MCL and MC unexpectedly closed - port=%2$s, state=%3$s

CWVTE2080W: Connection between MCL and EC unexpectedly closed

450
CWVTE2081W: Unable to find the requested resource file(%s)

Explanation: The requested resource file could not be found. User response: Ensure the requested file name and path are correct.
CWVTE2082W: Empty resource file(%s)

Explanation: The requested resource file is an empty file. User response: Ensure the requested file is a valid file.
CWVTE2083W: Failed to open requested resource file(%2$s), error=%3$s(%4$s)

CWVTE2084W: Failed on allocating memory for requested resource(%2$s), error=%3$s(%4$s)

CWVTE2085W: Failed on reading data from the requested resource(%2$s), error=%3$s(%4$s)

451
CWVTE2086W: Invalid URI(%s)

Explanation: The URI for the requested resource file is not valid. User response: Ensure the URI is valid that will be used for the requested resource.
CWVTE2087E: Error allocating memory for shut down detection

CWVTE2090E: Invalid port number

CWVTE2091W: Invalid trace level

CWVTE2092E: Invalid speak request params

Explanation: The specified speak parameters are invalid. User response: Supply valid parameters for the speak request.
Identifier
This section describes the identifiers for WebSphere Voice Server. WebSphere Voice Server and the MRCP client (such as the VoiceXML Browser) create unique identifiers in order to exchange information. You can see examples of these identifiers in the following sample SystemOut.log message:
452
CWVAE0318W: [sid=19009.IBM.9.22.96.50][rid=12][ltag=9507:9.22.96.33::4][turn=3] [pid=0x8e1e] Mailman::Dispatch: No audio available after 10 seconds. Forcing microphone off. [t=09:39:55:589 EDT]
[sid]
Stands for MRCP session-id. WebSphere Voice Server creates this unique identifier for the session and passes it to the MRCP client. In the example above, the [sid] is 19009.IBM.9.22.96.50.
[rid]
Stands for MRCP request-id. The MRCP client creates this unique identifier for the request and passes it to WebSphere Voice Server. In the example above, the [rid] is 12.
[ltag]
Stands for MRCP logging-tag. The MRCP client creates this unique identifier and passes it to WebSphere Voice Server. In the example above, the [ltag] is 9507:9.22.96.33::4.
[turn] Stands for the WebSphere Voice Server turn identifier. WebSphere Voice Server creates and manages this identifier. WebSphere Voice Server MRCP Synthesizer and Recognizer components count the number of Recognize or Speak MRCP messages. This identifier starts at 1. In the example above, the [turn] is 3. [pid] Stands for the WebSphere Voice Server process identifier. This identifier will be generated by platform OS. In the example above, the [pid] is 0x8e1e. [t] Stands for the WebSphere Voice Server event time stamp that is associated with exact time of event generation by originating component. WebSphere Voice Server engine components generate this time stamp as part of normal event logging. In the example above, [t] is 09:39:55:589 EDT. These identifiers allow for tracking the MRCP activity throughout the system on a per-session and per-request basis. If any of the four fields are not populated, then the information is not available at the time of the request. These identifiers are readily available to match up with the logs created by the MRCP client. Related information Message IDs This section describes the WebSphere Voice Server messages, which use a 10-character format.
Enabling and monitoring trace

Use trace to obtain detailed information about the execution of WebSphere Voice Server components. The trace logs for WebSphere Voice Server are located in one of the following paths based on your operating system: WAS_ROOT/logs/server1/ WAS_ROOT/logs/server1/ WAS_ROOT\logs\server1\
453
The trace log shows trace events such as function entries and exits, component events, and debug. Two topics provide additional information to guide system administrators: v Enabling trace v Monitoring trace
Enabling trace
Trace for a voice server process is enabled using the administrative console. You can configure the voice server to start in a trace-enabled state by setting the appropriate configuration properties. In the administrative console, enable and configure trace by using the following steps: 1. Expand Troubleshooting in the navigation panel by clicking the plus (+). 2. Click Logs and Trace. 3. Click the name of the server, such as server1, listed in the Server column of the table. 4. Click Diagnostic Trace. When you click the Modify button under the Trace Specification section of the Configuration tab, only three of the trace groups displayed in the Modify Trace panel are relevant to WebSphere Voice Server. They are the following: IBM WebSphere Voice Server Includes all of the basic WebSphere Voice Server components.
Table 19. WebSphere Voice Server components Acronym AdminConsole Common Config DTMFBean Dispatcher GramComp GramMgr LogTrace MediaConv RRBean RTSPBridge SesMgr Synth SysMgmt Description Administrative Console Common Component Configuration Subsystem DTMF Message Dispatcher Grammar Compiler Grammar Manager Log and Trace Subsystem Media Converter Resource Resolver RTSP Bridge Session Manager Synthesizer System Management
IBM WebSphere Voice Server ASR Includes all of the ASR components.
454
Table 20. ASR components Acronym ASRAPI ASRAUD ASRAW ASRBEAN ASRCONF ASRCTRL ASRDM ASRENG ASRFM ASRLAT ASRLM ASRNBST ASRSEQ ASRSIG ASRSRCH ASRTRN ASRVOC ASRWSI RECOGNIZER Description SMAPI Messages Audio Library Add Word ASR Bean Confidence Scoring Control Interface Detailed Match ASR Engine (configuration and recognition results) Fast Match Latency Language Model N-Best results Program Execution Sequence Signal Processing Stack Search Training Vocabulary and Pronunciations Work Station Interface Recognizer Bean
IBM WebSphere Voice Server TTS Includes all of the TTS components.
Table 21. TTS components Acronym CMM DIC ECI ETI MCL MLD MOR NOR PAR PHO PRE PRO SEM SOU SSM Description Concatenative Memory Manager Dictionary ECI Interface API Common Development Library Media and Control Layer Multilevel Data Structure Morphological Analyzer Normalizer Language Parser Phonetizer Preprocessor Prosody Semaphores Sound XML Parser, SSML Processor and Filter
455
Table 21. TTS components (continued) Acronym SYL TOR Description Syllable Analyzer Torrent
Important: Enable IBM WebSphere Voice Server ASR or IBM WebSphere Voice Server TTS tracing only if advised by IBM Support. Enabling these trace groups will cause performance issues. Because WebSphere Voice Server uses the common WebSphere approach to trace, the WebSphere Application Server Network Deployment 5.1.x Information Center contains detailed information on configuring trace in the administrative console. Trace Output Format, the last configuration parameter shown on the console, allows you to specify the level for trace output. Do not change this from Basic (Compatible).
Monitoring trace
WebSphere Voice Server uses the following log parameters: v Product = WVS v Manufacturer = IBM v LongName(ComponentId) = IBM WVS [component name] v ShortName(SourceId) = [component name]
Isolating audio problems between WebSphere Voice Server and the qualified IVR
Sometimes it is difficult to isolate audio problems between WebSphere Voice Server and the Interactive Voice Response (IVR). The first step toward problem determination is to ensure that the correct TTS audio is produced by the Voice Server. To do this, you can listen to the generated .pcm file for each TTS synthesis request. If the quality of the TTS audio is degraded from what you hear in the .pcm file, then the problem is most likely with the gateway. You can also listen to the .pcm files (using a tool of your choosing) for each ASR request to ensure that the correct ASR audio is produced. IBM WebSphere Voice Server MediaConv records the audio as it is received from the gateway. The ASR engine records pre- and post-endpointed audio. This allows you to hear if the ASR audio is degraded by the gateway. If entry/Exit or event is enabled for certain components, these components will log trace messages that reference binary or text files. For example, if event is enabled for the IBM WebSphere Voice Server MediaConv, then the Media Converter will create audio files to record all of the audio passed. On Linux and AIX, these additional files, including the .pcm files, are located in a WebSphere Voice Server-specific directory: WAS_ROOT/logs/server/CWV/timestamp/ For Windows Server 2003 Enterprise or Standard editions, the additional files, including the .pcm files, are located in a WebSphere Voice Server-specific directory: %WAS_ROOT%\logs\server\CWV\time-stamp\
456
Each time that WebSphere Voice Server is restarted, a new time-stamp directory is created so that no existing files are overwritten. For investigating audio problems, trace the following two components at the specified levels: 1. RTSPBridge at the all enabled trace level. 2. Media Converter (MediaConv) at the event trace level. Enabling entry/Exit or event for these components creates a lot of files on disk. You should only perform this activity at the direction of IBM Support or to investigate a specific audio problem. After you complete this activity, run the Collector Tool. The Collector Tool will also delete all information out of the CWV directory freeing up disk space. Administrative console Manage and view log and trace files and view alarms (console messages). The Voice Servers panel automatically refreshes every 60 seconds, but may be configured to refresh on a different schedule.
Table 22. Additional log files
Component Subcomponent Files created All utterances for all sessions are recorded. These recordings are endpointed audio. All utterances for all sessions are recorded. These recordings are not endpointed audio (that is, they contain extra silence). Saves data for site adaptation. SessionID.TurnID.ep.xxx (where xxx can be bsf, cep, or tag). File naming convention SessionID.TurnID.ep.pcm IBM WVS ASR - ASR If the trace level for engine subcomponent ASRAUD is set to entry/Exit, then post-endpointed audio is saved. IBM WVS ASR - ASR If the trace level for engine subcomponent ASRAPI is set to entry/exit, then pre-endpointed audio is saved.
SessionID.TurnID.pcm
IBM WVS ASR - ASR ASRTRN engine IBM WVS Media Converter If the trace level is set to event, then pre-endpointed audio is saved.
All audio for all SessionID.mc.asr.pcm sessions is recorded, meaning, or SessionID.TurnID.mc.asr.pcm the audio before it orSessionID.TurnID.mc.tts.pcm is passed to the ASR engine and the audio after it is received from the TTS engine.
Enabling trace for the Voice Trace Analyzer

The Voice Trace Analyzer in theWebSphere Voice Toolkit requires that you enable RTSPBridge tracing and set it to the all enabled trace level. To use the Voice Trace Analyzer to compute false accept (FA) or false reject (FR) accuracy, start the event handler for the ASRBEAN component and the entry and exit level tracing for the ASRAPI component. With all the options specified, RTSPBridge=all=enabled:ASRBEAN=event=enabled:ASRAPI=entryExit=enable displays in the trace specification modify window.
457
To use the Voice Trace Analyzer for audio analysis and playback, set the appropriate trace according to what is outlined in Table 21 on page 455, to obtain the audio files created during recognition processing. Remember: Run the collector tool to capture and erase the audio files so that your disk does not fill up. The resulting output JAR file from the collector tool can be imported into the Voice Trace Analyzer.
Collecting information for IBM Support

The collector tool gathers extensive information about your WebSphere Voice Server and WebSphere Application Server installation and packages it in a Java archive (JAR) file that you can send to IBM Customer Support to assist in problem determination. Information in the JAR file includes logs, property files, configuration files, operating system and Java data, and the absence or level of each software prerequisite. The collector tool runs to completion despite any errors that it might find, such as missing files or commands, and collects as much data in the JAR file as possible. Using the collector tool with the Voice Trace Analyzer Once you have collected all the trace information you need to debug a problem or analyze the recognition quality of your system, run the collector tool. You can import the resulting JAR file from the collector tool into the Voice Trace Analyzer. There are several steps to follow in order to run the collector tool. Run the collector command to create the JAR file if IBM Support needs more information to solve your problem.
Additional information
Detailed information about running the Collector tool may be found in the WebSphere Application Server Network Deployment 5.1.x Information Center.
Running the collector tool

Use the collector command to create the JAR file.
Steps for this task

1. Log on to the system as root if you are using AIX or Linux, or as Administrator if you are using a Windows platform. The collector tool should be run under the root or administrator user ID because some of the commands require system access authority. 2. Verify that Java 1.4.2 or higher is available in the path. The collector program requires Java code to run. Verify that all necessary information is in the path being used by the collector program and that you are not running the program from within the WebSphere Application Server product installation root directory. Verify that the path contains the following: v /bin v /sbin v /usr/bin v /usr/sbin
458
If your system is a Windows platform, include regedit in the path. 3. Create a work directory where you can start the collector program. 4. Make the work directory the current directory. The collector program writes its output Java archive (JAR) file to the current directory. The program also creates and deletes a number of temporary files in the current directory. Creating a work directory to run the collector program avoids naming collisions and makes cleanup easier. 5. Run the collector program by entering the collector command from the command line. Using the collector command with no additional parameters gathers one copy of the node data and data from each server in the node, and stores them in a single JAR output file. Note: Set the path correctly to use the non-qualified version of the command. Please note, WAS_ROOT/bin must be in the path to locate the collector command. The WebSphere Application Server installation root directory is determined at installation. It is identified in the setupCmdLine.sh file in Linux and AIX. On Windows, the installation directory was selected automatically when you installed WebSphere Voice Server. You can enter a fully qualified path to the collector command. For example, enter these commands in a default installation: On AIX:
>mkdir work >cd work >/usr/WebSphere/AppServer/bin/collector.sh
On Linux:
>mkdir work >cd work >/opt/WebSphere/AppServer/bin/collector.sh
On Windows:
C:\>md work C:\>cd work C:\>"C:\Program Files\WebSphere\AppServer\bin\collector"
The resulting JAR file

Name of the file The name of the JAR file is based on the host name and package of the Application Server product, in the format: hostname-ND-WASenv.jar or hostname-Base-WASenv.jar. For example, if you run the collector tool on the ws-laceweb server within a Network Deployment cell, the filename is ws-laceweb-ND-WASenv.jar . The Collector.log file is one of the files collected in the hostname-ND-WASenv.jar or hostname-Base-WASenv.jar file. Size of the file
459
The Collector Tool JAR size is determined by the following: v The collection of all properties files, configuration files, and operating system details v The SystemOut.log size and number of historical files v The trace.log size and number of historical files v The amount of audio files under the CWV directory Because the SystemOut.log and trace.log files are text-based, they compress well for inclusion in the Collector JAR. If you have the SystemOut.log file using the default values (a maximum size of 1 MB and one historical file) and a single 8 MB trace.log file, then your JAR size is 3.4 MB. If you have the SystemOut.log file using the default values (a maximum size of 1 MB and one historical file) and your trace.log file values set to 20 MB for the maximum file size with 19 historical files, then your JAR size is 95 MB. When the Collector Tool is run, all audio files are deleted from the CWV subdirectory.
What to do next
Send the hostname-ND-WASenv.jar or hostname-Base-WASenv.jar file to IBM Support for analysis.
Using IBM Support Assistant

The IBM Support Assistant helps you better search and find product support and education information. It also allows for the electronic creation and tracking of problem reports.
Installing IBM Support Assistant

Note: The IBM Support Assistant must be installed locally on each WebSphere Voice Server 5.1.x machine. 1. Install IBM Support Assistant version 2.0. Note: WebSphere Voice Server version 5.1.3 supports IBM Support Assistant version 2.0. 2. After you have installed IBM Support Assistant, install the WebSphere Application Server plug-in. 3. Then install the WebSphere Voice Server plug-in. Related information IBM Support Assistant: Product plug-in install, enable, disable and uninstall instructions
Troubleshooting by component: What is not working?

This section provides troubleshooting information based on the task you were trying to accomplish when the problem occurred. To find more information about your problem, select a task category from the list below.
460
If you do not see a task that resembles yours, or if the information provided does not solve your problem, contact IBM Support for further assistance.
Installation troubleshooting tips

If you are having problems installing the WebSphere Voice Server, follow these steps to resolve the problem. 1. If possible, follow the steps outlined in Troubleshooting the installation. 2. Browse the relevant log files for clues. v The main installation log file: $WVS_ROOT/log/CWVinstall.log $WVS_ROOT/log/CWVinstall.log %WVS_ROOT%\log\CWVinstallBase.log v IBM HTTP Server log: WAS_ROOT/logs/ihs_log.txt WAS_ROOT/logs/ihs_log.txt WAS_ROOT\logs\ihs_log.txt WAS_ROOT/logs/ihs_log.txt v IBM TTS log: $WVS_ROOT/IBM/ibmtts/log/install.log $WVS_ROOT/IBM/ibmtts/log/install.log %WVS_ROOT%\IBM\ibmtts\log\tts.log v WebSphere Application Server logs: WAS_ROOT/logs/server1 3. Ensure that you have installed the correct level of dependent software, such as the operating system version and revision level, by reviewing the Identifying hardware and software requirements section of the Information Center. If none of these steps solves the problem, check to see if the problem is identified and documented using the links in Diagnosing and fixing problems: Resources for learning. If you do not see a problem that resembles yours, or if the information provided does not solve your problem, contact IBM support for further assistance. For current information available from IBM Support on known problems and their resolution, see the IBM Support page. IBM Support has documents that can save you time gathering information needed to resolve this problem.
Administration and administrative console troubleshooting tips

This topic contains troubleshooting tips for administration and the administrative console.
Error creating ActionForm instance of class com.ibm.ws.console.wvs.form.ServerDetailForm

You can safely ignore the following error messages if they appear in a WebSphere Voice Server system log file. These errors do not cause a loss of data, nor do they impact the administrative use of the server or the speech server functions:
461
[9/15/05 9:23:59:062 MDT] 7b2b4323 WebGroup I SRVE0181I: [adminconsole] [/admin] [Servlet.LOG]: action: Error creating ActionForm instance of class com.ibm.ws.console.wvs.form.ServerDetailForm: java.lang.InstantiationException: com.ibm.ws.console.wvs.form.ServerDetailForm at java.lang.Class.newInstance3(Class.java(Compiled Code)) at java.lang.Class.newInstance(Class.java(Compiled Code)) at org.apache.struts.action.ActionServlet.processActionForm(ActionServlet.java:1706) at org.apache.struts.action.ActionServlet.process(ActionServlet.java:1563) at org.apache.struts.action.ActionServlet.doGet(ActionServlet.java:492) at javax.servlet.http.HttpServlet.service(HttpServlet.java:740) at javax.servlet.http.HttpServlet.service(HttpServlet.java:853) .... [9/15/05 9:23:59:219 MDT] 7b2b4323 WebGroup E SRVE0026E: [Servlet Error]-[action]: java.lang.NullPointerException at com.ibm.ws.console.wvs.action.ServerDetailAction.perform(ServerDetailAction.java:135) at org.apache.struts.action.ActionServlet.processActionPerform(ActionServlet.java:1791) at org.apache.struts.action.ActionServlet.process(ActionServlet.java:1586) at org.apache.struts.action.ActionServlet.doGet(ActionServlet.java:492) at javax.servlet.http.HttpServlet.service(HttpServlet.java:740) at javax.servlet.http.HttpServlet.service(HttpServlet.java:853) .... [9/15/05 9:23:59:250 MDT] 248c320 WebGroup E SRVE0026E: [Servlet Error]-[action]: java.lang.NullPointerException at com.ibm.ws.console.core.action.StatusAction.perform(StatusAction.java:56) at org.apache.struts.action.ActionServlet.processActionPerform(ActionServlet.java:1791) at org.apache.struts.action.ActionServlet.process(ActionServlet.java:1586) at org.apache.struts.action.ActionServlet.doGet(ActionServlet.java:492) at javax.servlet.http.HttpServlet.service(HttpServlet.java:740) at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
For detailed information and more tips, see the WebSphere Application Server Network Deployment 5.1.x Information Center.
Security troubleshooting tips

This topic contains troubleshooting tips for security components. Important: Java 2 Security must be disabled when using WebSphere Voice Server to avoid performance degradation. For detailed information, see the WebSphere Application Server Network Deployment 5.1.x Information Center.
IBM HTTP Server troubleshooting tips

This topic contains troubleshooting tips for the IBM HTTP Server. Note: This tip is specific to IBM HTTP Server on Windows. If you receive a Java FileNotFoundException error when using the IBM HTTP Server, use the following tip to resolve the problem. The IBM HTTP Server automatically uses AFPA directives to control the Fast Response Cache Accelerator. These directives must be disabled in order to remedy the error: 1. On your machine, locate the IBM HTTP Server installation directory. For example, C:\Program Files\IBMHttpServer. 2. In the /conf directory, locate the httpd.conf file, then perform the following steps: a. Save a backup copy of httpd.conf file. b. Edit the original httpd.conf file and comment out the following three AFPA lines (insert the pound symbol (#) in front of each line):
# AfpaEnable # AfpaCache on # AfpaLogFile "path/to/afpalog" V-ECLF
c. Add a Listen 80 directive in the httpd.conf file.
462
The Listen 80 directive is needed because port 80 only binds the root process to the port. By default on a Windows system, the IBM HTTP Server listens on the AFPA socket and not on the default Windows socket. Adding the Listen 80 directive forces the Web server to listen on the default Windows socket when AFPA is disabled. 3. Stop the IBM HTTP Server. 4. Start the IBM HTTP Server.
TTS troubleshooting tips

This topic contains troubleshooting tips for text-to-speech (TTS). v The TTS engine(s) will not start (Linux platforms) v Speech Synthesis Markup Language or lexicon-related troubleshooting tips v The embedded audio or lexicon document cannot be fetched, or the embedded audio file within the SSML document will not play on page 464 v Audio specified in the SSML <audio> tag not heard on page 464 v The SSML document encounters a parse error on page 464 v When the <voice> tag has a gender attribute set to neutral, the voice is not changed on page 464 v A synthesized audio output is truncated on page 464 v v v v v v Prosody volume and range attributes have no effect on page 465 Age and variant voice settings unsupported in the TTS on page 465 Australian TTS and city name pronunciations on page 465 Speech markup <emphasis> element limitation on page 465 Errors when trying to load multiple voices on page 465 Reliability issues when antivirus software installed on page 466
The TTS engine(s) will not start (Linux platforms)

Use the following checklist to troubleshoot this problem: v Run the following command: rpm -qa | grep ibmtts The language you are trying to use should be installed. v Make sure you have installed the correct language. To do this, navigate to /var/opt/IBM/ibmtts/cfg/eci.ini and look for the following entries: [1.0], [1.1], [8.0], [6.0], [3.1], [4.0], [2.1]. The [1.0] is for American English, the [1.1] is for British English, the [8.0] is for Japanese, the [6.0] is for Chinese, [3.1] is for Canadian French, and [4.0] is for German, [2.1] is for Latin American Spanish. v Check the WAS_ROOT/logs/server1/SystemOut.log for errors.
Speech Synthesis Markup Language or lexicon-related troubleshooting tips

v Is your Speech Synthesis Markup Language (SSML) compliant with SSML 1.0 standards? v Have you validated your lexicon file using the Voice Toolkit Lexicon Editor? For more information on the Lexicon Editor provided in WebSphere Voice Toolkit, see Validating a lexicon file in the Voice Toolkit Information Center. v Check the WAS_ROOT/logs/server1/SystemOut.log to see if any errors have been logged.
463
The embedded audio or lexicon document cannot be fetched, or the embedded audio file within the SSML document will not play
v Make certain that the IBM HTTP Server is up and running on the host system: 1. Using Linux and AIX, change directory to /opt/IBMHttpServer/bin and type ./apachectl start. 2. Then check to make sure that the server is started by running the following command: ps auxw | grep httpd. 3. On a Windows platform, view your services panel to check if the IBM HTTP Server is running. v Is the path you specified correct for the file you want? v Check the WAS_ROOT/logs/server1/SystemOut.log to see if any errors have been logged. v Does the file exist in the location specified? v Verify that the file is in the correct audio format. VoiceXML specifications require that a platform support the following playing and recording audio formats: Raw (headerless) 8kHz 8-bit mono mu-law [PCM] single channel. Raw (headerless) 8kHz 8 bit mono A-law [PCM] single channel. WAV (RIFF header) 8kHz 8-bit mono mu-law [PCM] single channel. WAV (RIFF header) 8kHz 8-bit mono A-law [PCM] single channel. Also, if the SSML processor cannot find the audio file, it will synthesize the files that it can find. This means that there is no error message rendered, but you will not hear the file played. In order to troubleshoot this error you will need to turn on tracing to see why the file was not fetched.
Audio specified in the SSML <audio> tag not heard

If you are unable to hear the audio specified in the SSML <audio> tag, verify that the server where WebSphere Voice Server is installed, doesnt have Windows Internet Information Services (IIS) or any other web servers running or installed. Disable any web server software you find running or installed as additional Web servers conflict with the WebSphere Application Server HTTP server.
The SSML document encounters a parse error

v Is your SSML compliant with SSML 1.0 standards? v Have you validated your lexicon file using the Voice Toolkit Lexicon Editor? For more information on the Lexicon Editor provided in WebSphere Voice Toolkit, see Validating a lexicon file in the Voice Toolkit Information Center. v Check the WAS_ROOT/logs/server1/SystemOut.log to see if any errors have been logged.
When the <voice> tag has a gender attribute set to neutral, the voice is not changed
Certain <voice> tag attributes, such as gender neutral and age, are not supported for any concatentive voices, including US English and Chinese voices. Also note that the variant attribute is not supported for the US English and Chinese voices.
A synthesized audio output is truncated
464
The amount of text included in one speak request is limited to the amount of text that can be spoken in five minutes. The synthesized audio for larger amounts of text may be truncated.
Prosody volume and range attributes have no effect

The prosody volume and range attributes do not have the appropriate effect on the TTS generated and are currently unsupported.
Age and variant voice settings unsupported in the TTS

The TTS supports only adult voices. Some age and variant voice settings may not affect the TTS.
Australian TTS and city name pronunciations

The Australian TTS uses the UK English TTS. As a result, some of the names of Australian cities will sound as if pronounced in UK English. You may insert lexicons into the VoiceXML script to correct the pronunciations.
Speech markup <emphasis> element limitation

The <emphasis> element, included in VoiceXML 2.0, is currently not supported. A fix is planned for a future release.
Errors when trying to load multiple voices

AIX When loading more than four different TTS voices, such as the U.S. English male and female voices, the U.K. male and female voices, and the Latin American Spanish male and female voices, you may need to increase the maximum number of shared memory segments available to TTS. To do this, use the following instructions: 1. Add export EXTSHM=ON to the WebSphere Voice Server Tuning Updates section of the /etc/profile. 2. Shut down WebSphere Application Server. 3. Log off from the machine. 4. Log in to the machine as root. 5. Restart WebSphere Application Server. Windows When loading more than three different TTS voices, for example the U.S. English male and female voices, as well as the U.K. English male and female voices, you may need to boot Windows Server 2003 so that it allocates more contiguous memory to application programs. To do this, use the following instructions: Important: Allocating more contiguous memory may degrade the performance of some systems, so only use this procedure if necessary. See Memory Support and Windows Operating Systems for more information. Restriction: This procedure only works on machines with at least 1 GB of memory running Windows Server 2003 Standard or Enterprise editions.
465
1. Change directory to your Windows boot partition.

C: cd \
2. At the command line prompt, type: attrib boot.ini You should see a response similar to the following: A SH 3. Next, type:
attrib -h -s boot.ini notepad boot.ini
C:\boot.ini
You should see something similar to the following in Notepad:

[boot loader] timeout=30 default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS [operating systems] multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Windows Server 2003, Enterprise" /fastdetect
4. Carefully duplicate and edit the last line, adding /3GB as in the following, and save the file:
[boot loader] timeout=30 default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS [operating systems] multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Windows Server 2003, Enterprise /3GB" /fastdetect /3GB multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Windows Server 2003, Enterprise" /fastdetect
5. Reboot your machine, and when given the choice, select Windows Server 2003, Enterprise /3GB, or wait until the machine automatically boots the new default /3GB configuration. This configuration allows you to run additional voices, given that you have sufficient real memory and/or pagespace. If there is any problem with the /3GB configuration, you can boot from the original configuration (the second line on the boot screen) and remove the /3GB line from boot.ini. To hide the boot.ini file again, type: attrib +h +s boot.ini
Reliability issues when antivirus software installed

If you have antivirus software installed or enabled on a WebSphere Voice Server Windows Server 2003 machine, you may encounter the following: v Performance issues v ASR and TTS engines time out v Windows Server 2003 stops If the operating system stops, the Windows Server 2003 Online Crash Analysis will indicate that the antivirus software has had a critical failure and request that you update the antivirus software to correct the problem. Also, it is possible that the antivirus scanning software is attempting to scan some or all of the WebSphere Voice Server files; therefore, it is recommended that you disable scanning for the following directories (where installation_drive is the letter designation of the drive where the software is installed, such as C:): v %WVS_ROOT% (the directory where WebSphere Voice Server is installed) v WAS_ROOT (the directory where WebSphere Application Server is installed) v installation_drive\Program Files\IBMHttpServer (the directory where IBM HTTP Server is installed) v The directory defined by the TEMP environmental variable (%TEMP%) v The directory defined by the TMP environmental variable (%TMP%)
466
If you have configured WebSphere Voice Server to point the General (first level) Cache and Grammar (second level) Cache to directories other than the default, then you must disable those directories as well. If scanning on the above directories is disabled, and problems are still occurring, it may be necessary to uninstall or disable the antivirus software. Updating the antivirus software may not correct the problems. If not, contact your antivirus software vendor for a solution.
ASR troubleshooting tips

This topic contains troubleshooting tips for automatic speech recognition (ASR).
General
v The ASR engine(s) will not start v Reliability issues when antivirus software installed on page 468
Grammar or lexicon
v A grammar has compilation errors on page 468 v A grammar cannot compile with special characters on page 468 v A grammar or lexicon cannot be fetched on page 469 v Compilation of a large grammar or a large lexicon is slow and times out on page 469 v An external lexicon, grammar, or audio resource has been updated but WebSphere Voice Server is using an old copy on page 469 v MIME type and universal grammars on Genesys on page 470 v The system cannot find the path specified error on page 470 v Compilation of a large grammar fails with CWVGC2041E on page 470
Recognition
v v v v v v v The environment is noisy on page 470 Large grammar concerns on page 471 No-input timeout is happening too often on page 472 No-match is happening too often on page 472 Increasing the accuracy of my grammar on page 472 False barge-ins are occurring on page 472 Circular references in SRGS semantic interpretation results on page 472
The ASR engine(s) will not start

This problem could be caused by a couple of issues. The ASR engines will not start if they cannot load their dynamic libraries. Linux will not be able to load the libraries if the LD_LIBRARY_PATH is not properly set. Similarly, AIX will not load the libraries if the LIBPATH is not set properly. If you are using WebSphere Voice Server versions 5.1, 5.1.1, or 5.1.2 on Linux this path is set when you run the wvs.setenv script. If you are using WebSphere Voice Server version 5.1.3 on Linux or AIX, the installation adds the settings contained in the wvs.setenv script to the /etc/profile file. On Windows, the path to the libraries is set up by the installation process.
467
See the Tuning section for more information about the modifications that take place during the installation process. If this information does not solve your problem, see the IBM Support topic for further assistance.
Reliability issues when antivirus software installed

If you have antivirus software installed or enabled on a WebSphere Voice Server Windows Server 2003 machine, you may encounter the following: v Performance issues v ASR and TTS engines time out v Windows Server 2003 stops If the operating system stops, the Windows Server 2003 Online Crash Analysis will indicate that the antivirus software has had a critical failure and request that you update the antivirus software to correct the problem. Also, it is possible that the antivirus scanning software is attempting to scan some or all of the WebSphere Voice Server files; therefore, it is recommended that you disable scanning for the following directories (where installation_drive is the letter designation of the drive where the software is installed, such as C:): v %WVS_ROOT% (the directory where WebSphere Voice Server is installed) v WAS_ROOT (the directory where WebSphere Application Server is installed) v installation_drive\Program Files\IBMHttpServer (the directory where IBM HTTP Server is installed) v The directory defined by the TEMP environmental variable (%TEMP%) v The directory defined by the TMP environmental variable (%TMP%) If you have configured WebSphere Voice Server to point the General (first level) Cache and Grammar (second level) Cache to directories other than the default, then you must disable those directories as well. If scanning on the above directories is disabled, and problems are still occurring, it may be necessary to uninstall or disable the antivirus software. Updating the antivirus software may not correct the problems. If not, contact your antivirus software vendor for a solution.
A grammar has compilation errors

Use the WebSphere Voice Toolkit or the command-line voiceTest tool to extract the compilation messages from the package file. If the messages indicate that the problem is caused by missing baseforms, then use the Voice Toolkit to generate the appropriate baseforms using the lexicon editor. Refer to the WebSphere Voice Toolkit Information Center for details.
A grammar cannot compile with special characters

You may receive syntax or semantic errors when using Japanese and other special characters in your grammar. Your trace logs will read:
failed-uri-cause: Unable to compile:session:grammar101:Invalid Characters in Root Rule Name completion-cause: 005 gram-comp-failure failed-uri: session:grammar101
468
Remove any illegal characters from your grammar and recompile. Refer to section 3.1 of the SRGS specification for additional details.
A grammar or lexicon cannot be fetched

Use a browser to check make sure that the grammar or lexicon can be viewed successfully. Also be sure to check for typos in the grammar or lexicon uniform resource identifier (URI).
Compilation of a large grammar or a large lexicon is slow and times out

Use the WebSphere Voice Toolkit or the command-line voiceTest tool to see if the file has a lot of auto-added pronunciations. If so, add these pronunciations to the lexicon. You can preload any large grammars using the WebSphere Voice Server administrative console. Select the configuration tab and use the ASR Configurations panel to add the fully qualified path name to the grammar you want preloaded. Click OK.
An external lexicon, grammar, or audio resource has been updated but WebSphere Voice Server is using an old copy
WebSphere Voice Server uses HTTP caching parameters to expire the cached resource in the general (first level) cache. A cached resource can be an external lexicon, grammar or audio file. The expiration for a cached resource can be set on the Web server, application server, or specified by the content of the lexicon or grammar using the meta tag <meta http-equiv="Expires" content="0"></meta>. If the expiration is not provided by the fetched resource, then the default expiration time is used. For example, the default expiration might be set to expire after 24 hours. The default expiration time is specified by the com.ibm.voice.server.rr.fetchexpires configuration parameter. During application development for grammar, lexicon, or audio files, disable the general cache, by setting com.ibm.voice.server.rr.caching=false. This ensures that WebSphere Voice Server fetches the most current content every time. Once application development is finished, apply the appropriate caching rules to the resource according to the VoiceXML and HTTP specifications, and enable the general cache. Note: Setting this configuration parameter to true, com.ibm.voice.server.rr.caching=true enables the general cache. If you need to delete the WebSphere Voice Server general and grammar (second level) caches, perform the following steps: 1. Shut down WebSphere Voice Server. 2. Go to the directory containing the WebSphere Voice Server general cache. This is specified by com.ibm.voice.server.rr.cachename. If you have not modified this configuration parameter, then it is $WVS_ROOT/rrcache. 3. Remove all of the files so that the entire directory is empty. 4. Go to the directory containing the WebSphere Voice Server grammar cache. This is specified by com.ibm.voice.server.gm.cachename. If you have not modified this configuration parameter, then it is $WVS_ROOT/gmcache. 5. Remove all of the files so that the entire directory is empty. 6. Restart WebSphere Voice Server.
469
See the Configuration parameters on page 105 page for more information on all of the cache-related parameters.
MIME type and universal grammars on Genesys

Note: The information in this section does not apply to WebSphere Voice Server version 5.1.3. When using universal grammars, such as help, cancel and exit, on a Genesys Interactive Voice Response (IVR) system, you must change the Multipurpose Internet Mail Extensions (MIME) type of these grammars to application/srgs+xml, or you will receive an error detected message. Genesys ships the universal grammars (.grxml) with a MIME type of text/plain. WebSphere Voice Server expects files with the .grxml extension to have a MIME type of application/srgs+xml. For more information on the supported MIME types in WebSphere Voice Server, refer to Configuring a MIME map on your customer Web server on page 105.
The system cannot find the path specified error

If you are using Windows Server 2003 and a grammar or lexicon fetch returns with the CWVRR5004E error of The system cannot find the path specified, then the problem may be that the grammar or lexicon cache entry is too long. On Window Server 2003, the maximum length of a fully qualified file name is 255 characters. If a computed grammar or lexicon cache entry URI is longer than 255 characters, then the grammar or lexicon cannot be added to the cache. To resolve this issue, do one of the following: v Shorten the URI reference until the cache entry is less than 255 characters. v Make the grammar or lexicon non-cacheable by modifying the cache control to disable caching. v Shorten the name of the WebSphere Voice Server general and grammar (second level) cache directory locations. This is specified by com.ibm.voice.server.rr.cachename and com.ibm.voice.server.gm.cachename. See the Configuration parameters on page 105 page for more information on all of the cache-related parameters.
Compilation of a large grammar fails with CWVGC2041E

The error message (CWVGC2041E) occurs because of deep recursive calls in the Grammar Compiler, such as when the Grammar Compiler tries to compile a grammar with a rule with very long sequential tokens. A protection mechanism has been added to the Grammar Compiler to catch the stack overflow and report the error message (CWVGC2041E). If users see the error message, they can reduce the rule length to avoid the core dump. Since a lengthy rule is usually not likely in a production application grammar, the long rule error might be caused by a typo or badly written grammar.
The environment is noisy

Adjust the sensitivity level. Lower the level if the background noise/speech is being recognized or causing barge-ins.
470
Large grammar concerns

Use the WebSphere Voice Toolkit or the command-line voiceTest tool to generate a lexicon and see if any words are being auto-added. Make sure that the spellings and pronunciations are correct. Also see if the grammar accepts too many phrases (use fsgenum to list acceptable phrases). There is a limit of one million pronunciations per grammar, so you may need to keep to fewer than 500,000 different words. This limit is dependent upon the language and vocabulary you are using. If a grammar is not compiling, and you have tracing turned on, you may see an out of memory error similar to the following in the trace.log:
[6/22/04 13:49:56:702 EDT] 257e78a2 ExceptionUtil E CNTR0020E: Non-application exception occurred while processing method "defineGrammar" on bean "BeanId(WVS5.1-kirkwood#wvsejb.jar#RecognizerEJB, null)". Exception data: java.lang.OutOfMemoryError [6/22/04 13:49:56:820 EDT] 257e78a2 LocalTranCoor E WLTC0017E: Resources rolled back due to setRollbackOnly() being called. [6/22/04 13:49:56:881 EDT] 257e78a2 Dispatcher E CWVMD0003E: [sid=0.IBM.34.34.130.1][rid=100][turn=1] Failed to process recognizer request. Reason: ; nested exception is: com.ibm.ws.exception.WsEJBException: nested exception is: java.lang.OutOfMemoryError
Or, if you are trying to compile a grammar using voiceTest and it takes 10 minutes or more to compile, you may see an error similar to the following:
define -grammar /opt/WebSphere/VoiceServer/lib/busLssi.bnf IBM WebSphere Voice Server voiceTest IBM WebSphere Voice Server voiceTest [04/06/22 13:37:06:928 EDT] main pt A CWVPT0027I: Connecting to localhost 554... [04/06/22 13:37:07:226 EDT] Thread-1 pt A CWVPT0029I: Waiting for reply... [04/06/22 13:37:07:227 EDT] Thread-1 pt A CWVPT0030I: RTSP session: "0.IBM.34.34.130.1" [04/06/22 13:37:07:230 EDT] main pt A CWVPT0028I: RTP contact at 34.34.130.1 50002 [04/06/22 13:49:56:904 EDT] main pt E CWVPT0010E: RTSP error 500 Internal Server Error
These errors indicate that you need to increase the WebSphere Application Server Maximum Heap Size setting. The default setting is 256 MB for WebSphere Voice Server versions 5.1, 5.1.1, and 5.1.2. The default setting for WebSphere Voice Server version 5.1.3 is 512 MB. To 1. 2. 3. 4. increase the Maximum Heap Size setting, do the following: Launch and log in to the administrative console. In the left navigation pane, select Servers, and then select Application Servers. Click the name of the server you are using. By default, it is server1. On the server panel, in the Additional Properties table, click Process Definition. 5. On the Process Definition panel, in the Additional Properties table, click Java Virtual Machine. 6. On the Java Virtual Machine panel, in the Configuration table, increase the value of the Maximum Heap Size in MB. 7. Click the Apply button at the bottom of the Configuration table to save your changes to the master configuration.
471
8. Restart WebSphere Application Server for the changes to take effect.
No-input timeout is happening too often

Check audio level. Raise the audio gain if it is low, or raise the sensitivity level.
No-match is happening too often

Use the following checklist to troubleshoot this problem: v Make sure the audio matches the pronunciations of the words in the grammar. v Make sure that the timeouts are not too short. If so, then a phrase may be truncated by a short pause. v Make sure that audio gain is not too high. If so, then the signal is clipped. v Make sure that audio is not distorted by the voice activity detection in the gateway (audio with perfect artificial silence). v Adjust the speed-accuracy knob towards the accurate level (but note that this will impact speed performance).
Increasing the accuracy of my grammar

Use the following checklist to troubleshoot this problem: v Check the spelling of your words and minimize use of abbreviations when possible. For example, use Street instead of St. and Road instead of Rd. v Break long compound words or numbers into separate words if are difficult to say without pausing. v Check the coverage of the grammar. For example, listen to recordings of live interactions and make sure the grammar covers what is being said. v Remove nonsensical or low-probability phrases that can cause confusion. Enumerate all phrases the grammar will accept and redesign if there are too many. v Adjust the weights of rules. See the Speech Recognition Grammar Specification (SRGS) 1.0 at http://www.w3.org/TR/speech-grammar/.
False barge-ins are occurring

False barge-ins may be caused by loud breaths, clicks, or by the prompt itself, if the echo has not been suppressed. Use the following checklist to troubleshoot this problem: v Make sure the echo cancellation is activated correctly. v Lower the sensitivity level. v Reduce the volume of the prompt.
Circular references in SRGS semantic interpretation results

SRGS semantic interpretation results may erroneously include references to objects that contain circular references. In this case, the circular reference would cause the NLSML generation to recurse indefinitely. The Grammar Compiler resolves this erroneous SISR by limiting the stack depth to 20 properties. A semantic result cannot contain a tree of properties deeper than 20 nodes.
472
IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server troubleshooting tips
This topic contains troubleshooting tips for the IBM Text-to-Speech (TTS) Connector for Genesys Voice Platform (GVP) for WebSphere Voice Server. If you are experiencing problems with the IBM TTS Connector, enable trace for the IBM TTS Connector. Three parameters control the logging and tracing in the IBMTextToSpeech.ini: EnableTrace, MrcpTraceDir, and LogFile. To enable trace on the TTS Connector, set the EnableTrace parameter in the IBMTextToSpeech.ini file to 1 instead of 0 (EnableTrace=1). The log file is located at LogFile=LOGDIR\mrcplog.txt, where LOGDIR is the directory for log and trace files. The installer sets this to the directory where Genesys puts its log files, however you may wish to change it because the file has a tendency to grow quite large on systems with a lot of calls. By default (EnableTrace=0), only errors are logged in this file. MrcpTraceDir=LOGDIR is the directory where the Media Resource Control Protocol (MRCP) trace files are created. A separate file is created for each channel and will contain all the messages exchanged through the MRCP connection for that channel. The files will be named in the MRCP-xxx.txt convention, where xxx is the channel number. These files will be created only if EnableTrace=1. For more information on IBMTextToSpeech.ini file parameters, refer to Configuring the IBM TTS Connector for Genesys Voice Platform for WebSphere Voice Server.
IBM TTS Connector error codes and messages

MMMM: Explanation: The first line of the Real Time Streaming Protocol (RTSP) or MRCP reply message or the non-zero value in the completion-code header. User Response: Consult the WebSphere Voice Server text-to-speech documentation for further details. No channel available: Explanation: Time out occurred waiting for a free channel on Genesys or no TTS engine was available in WebSphere Voice Server. User Response: Consult the Genesys and WebSphere Voice Server configuration documentation on how to increase the number of TTS engines available. Ensure that the IBMTextToSpeech.ini and WebSphere Voice Server have the same number of TTS engines. Connect error: X (D) Explanation: Socket connection error. X and D contain the socket error code in hex and decimal. A TCP socket connection failed. This indicates a network or a setup problem (unreachable or unknown RTSP host for example). User Response: Type net helpmsg dddd, where dddd is the error code, at the command line. recv error: X (D) Explanation: Socket read error. X and D contain the socket error code in hex and decimal. This error indicates a network communication problem between the IBM TTS Connector and the WebSphere Voice Server.
473
User Response: Type net helpmsg dddd, where dddd is the error code, at the command line. RTSP error: MMMM Explanation: MMMM contains the first line of the RTSP message. User Response: Type net helpmsg dddd, where dddd is the error code, at the command line. You can also examine the MRCP-xxx trace files for details. Malformed RTSP message: LLL Explanation: Could not parse the RTSP line, where LLL is the line. This indicates a network communication problem, like connecting to a non MRCP socket. User Response: Check the RTSP URL in the configuration file and examine the MRPC-xxx trace files. Malformed RTSP header: HHH Explanation: Could not parse the RTSP header. HHH contains the line with the invalid header. User Response: Check the MRCP-xxx trace files for more information. Could not setup synthesizer service Explanation: RTSP SETUP failed. This error is returned when a SETUP request fails for any reason other than an RTSP or network communication error. User Response: Check the MRCP-xxx trace files for more information. MRCP reply not expected: [MMMM] Explanation: MMMM contains the first line of the unexpected MRCP message. User Response: Check the MRCP-xxx trace files for more information.
Gateway communication problems

This topic contains troubleshooting tips for communication problems for gateways. Ensure that the following RTSP parameters are set correctly in the gateway: v Port configuration parameter v Synthesizer media parameter v Recognizer media parameter
voiceTest troubleshooting tips

This topic contains troubleshooting tips for the voiceTest tool.
MRCP failed URI error message

If your grammar does not have a language attribute and you are using voiceTest, you may receive the following MRCP error:
MRCP/1.0 101 407 COMPLETE failed-uri-cause: Speech grammar must specify language completion-cause: 004 gram-load-failure failed-uri: session:grammar101
To resolve this issue, specify the language attribute either in the grammar file or by using the voiceTest -speech-language command-line parameter.
474
Troubleshooting by task: What are you trying to do?

This section provides troubleshooting information based on the task you were trying to accomplish when the problem occurred. To find more information about your problem, select a task category from the following list. If you do not see a task that resembles yours, or if the information provided does not solve your problem, contact IBM Support for further assistance.
Troubleshooting installation problems

Select the problem you are having with the WebSphere Voice Server installation: v Installation either completes with errors or warnings, or hangs. v The installation process completes, but the application server or deployment manager does not start or starts with errors. v The installation completes, but the WebSphere Voice Server installation verification does not work. v There are failures or error messages issued during installation. v The installation completes but the administrative console does not start. If you do not see a problem that resembles yours, or if the information provided does not solve your problem, see Troubleshooting the Installation. For current information available from IBM Support on known problems and their resolution, see the IBM Support page. IBM Support has documents that can save you time gathering information needed to resolve this problem. Before opening a PMR, see the IBM Support page.
Failures and errors while installing the product

Failures on Red Hat Linux
If your installation of WebSphere Voice Server, text-to-speech (TTS) languages, or automatic speech recognition (ASR) languages fails on a Red Hat Linux platform, your machines Red Hat package manager (RPM) database, which stores system installation information, may be corrupted. To check your machine, use the following steps: 1. Issue the rpm -qa | grep IBM command. 2. If you see error messages from RPM database after issuing the rpm -qa | grep IBM command, check the /var/opt/IBM/ibmtts/log/install.log file or the /tmp/CWVOptInstall.log file to see if the file contains error information similar to the following:
wrn, The installer could not successfully add the product information into the RPM database. Installation will continue as this is not critical to the installation of the product.
If you have a similar error message, then your machines RPM database has been corrupted. This generally occurs if the machine is powered off midway through installing an application. This error has also occurred on systems with bad hard disk sectors. To fix the problem:
475
1. Uninstall WebSphere Voice Server. 2. From a system shell running as a root user, issue the rpm --rebuilddb command. 3. Issue the rpm -qa command again to confirm that there are no error messages generated. 4. Reinstall WebSphere Voice Server.
Errors
If the WebSphere Voice Server installation program indicates that errors were encountered while installing the product, do the following: 1. Browse all of the log files in $WVS_ROOT/logs/ for clues. Pay particular attention to the main installation log file, CWVinstall.log. 2. Check the command prompt from which the installation panel was launched for error messages. 3. Look up any error or warning messages in the message reference table. 4. Review Troubleshooting the Installation.
Installation completes but IVT does not work

If the WebSphere Voice Server installation program completes successfully, but the Installation Verification Test (IVT) does not run, do the following: 1. Browse the application server log files $WVS_ROOT/logs/server_name/ SystemErr.log and SystemOut.log for any indication as to what might be causing the IVT issue. 2. If the ASR and or the TTS engine does not start, the IVT will fail. Check the server logs from step one and ensure these engines started. 3. View the JVM logs of the hosting application server for clues, after attempting to run WebSphere Voice Server IVT. 4. Look up any error or warning messages in the message reference table. 5. Verify any security-related problems. 6. Verify that $SPCH_ROOT is set. If it is not, then it is likely that wvs.setenv script was not run. Note: This step applies only to WebSphere Voice Server versions 5.1, 5.1.1, and 5.1.2. Using WebSphere Voice Server 5.1.3, the wvs.setenv script runs automatically. 7. Review Troubleshooting the Installation for more information. If you do not see a problem that resembles yours, or if the information provided does not solve your problem, check to see if your problem has been identified. See Diagnosing and fixing problems: Resources for learning for information on getting the latest updates. If your problem has not been reported, see the IBM Support page for further assistance.
Troubleshooting testing and first time run problems

Select the problem you are having with testing WebSphere Voice Server: v The application server or deployment manager does not start or starts with errors. v The application does not start or starts with errors. v There are errors after enabling security. v There are access problems after enabling security.
476
v There are errors after enabling Secure Sockets Layer. v There are Errors in messaging. If you do not see a problem that resembles yours, or if the information provided does not solve your problem, see the IBM Support page. The IBM Support site provides current information available on known problems and their resolutions. IBM Support has documents that can save you time gathering information needed to resolve this problem. Before opening a Problem Management Report (PMR), see the IBM Support page.
Troubleshooting application runtime and management problems

This topic contains troubleshooting tips for application runtime and management problems. For detailed information, see the WebSphere Application Server Network Deployment 5.1.x Information Center.
477
478
Chapter 9. Resources for learning

Use the following links to learn more about the technologies related to IBM WebSphere Voice Server: Operating systems: v IBM AIX 5L Web site v Microsoft Windows Server 2003 Web site v Red Hat Web site v SUSE LINUX Web site Related products and specifications: v WebSphere Application Server zone v Sun Microsystems Web site v Java specification documents
479
480
Chapter 10. Notices and trademarks

Copyright notice
Copyright IBM Corporation 2004, 2005. All rights reserved. May only be used pursuant to an IBM software license agreement. No part of this publication may be reproduced, transmitted, transcribed, stored in a retrieval system, or translated into any computer language, in any form or by any means, electronic, mechanical, magnetic, optical, chemical, manual, or otherwise, without prior written permission of IBM Corporation. IBM Corporation grants you limited permission to make hardcopy or other reproductions of any machine-readable documentation for your own use, provided that each such reproduction shall carry the IBM Corporation copyright notice. No other rights under copyright are granted without prior written permission of IBM Corporation. The document is not intended for production and is furnished as is without warranty of any kind. All warranties on this document are hereby disclaimed, including the warrenty of non-infringement and the implied warranties of merchantability and fitness for a particular purpose. U.S. Government Users Restricted RightsUse, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corporation.
Trademarks
Advanced Function Presentation, AFP, AIX, AIX 5L, DB2, developerWorks, Domino, e-business on demand, Everyplace, IBM, Lotus, Lotus Notes, POWER, pSeries, Rational, Redbooks, Tivoli, ViaVoice, WebSphere, xSeries, and z/OS are trademarks or registered trademarks of the International Business Machines Corporation in the United States, other countries, or both. Intel, Intel Inside (logos), MMX and Pentium are trademarks of Intel Corporation in the United States, other countries, or both. Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both. Linux is a trademark of Linus Torvalds in the United States, other countries, or both. Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both. UNIX is a registered trademark of The Open Group in the United States and other countries. Other company, product, and service names may be trademarks or service marks of others.
481
482
Glossary
2PKDP. See Two-Party Key Distribution Protocol. 3G. See third generation. ADSL. See Asymmetric Digital Subscriber Line. Advanced Function Presentation. A set of licensed programs, together with user applications, that use the all-points-addressable concept to print data on a wide variety of printers or to display data on a variety of display devices. AFP includes creating, formatting, archiving, retrieving, viewing, distributing, and printing information. Advanced Interactive eXecutive. See AIX. Advanced Mobile Phone Service (AMPS). The original analog cellular network and currently the worlds largest cellular standard. Advanced Radio Data Information Service (ARDIS). A wireless data network from American Mobile Satellite Corporation that covers more than 11,000 cities and towns in the U.S. Operating in the 806-824/851-869MHz bands, ARDIS is a packet-switched network that provides a data rate of 19.2 Kbps and is known for its deep penetration into buildings. Primarily used for field service and transportation applications, ARDIS was created by Motorola in the mid-1980s for IBMs field service division and was later spun off as a commercial service. AIX (Advanced Interactive eXecutive). A UNIX operating system developed by IBM that is designed and optimized to run on POWER microprocessor-based hardware such as servers, workstations, and blades. AMI-C. See Automotive Multimedia Interface-Collaboration. AMPS. See Advanced Mobile Phone Service. ARDIS. See Advanced Radio Data Information Service. Asymmetric Digital Subscriber Line (ADSL). A technology based on Digital Subscriber Line (DSL) technology that is for Internet access where fast downstream is required, but slow upstream is acceptable. ASYNC. See asynchronous. asynchronous (ASYNC). Pertaining to events that are not synchronized in time or do not occur in regular or predictable time intervals. For example, input events are controlled by the user; the program can read them later.
asynchronous transfer mode (ATM). A communications method in which information is organized into small, fixed-length cells that recur without regular or predictable time relationships. ATM incorporates network switches and negotiated service connections. ATM. See asynchronous transfer mode. Automotive Multimedia Interface-Collaboration (AMI-C). A worldwide organization of automobile manufacturers that is involved in the standardization of automotive multimedia interfaces to motor vehicle communication networks. base station (BS). An earth-based transmitting and receiving station for cellular phones, paging services and other wireless transmission systems. Bluetooth. Technology specification created for short-range (up to 10 meters) wireless connection using low-cost transceiver chips to be embedded in mobile PCs, smart phones, and other portable devices. Provides three voice and data channels via a one-to-one connection with built-in encryption and verification. BOOTP. See Bootstrap Protocol. Bootstrap Protocol (BOOTP). A protocol that allows a client to find both its Internet Protocol (IP) address and the name of a file from a server on the network. bot. A program used on the Internet that performs a repetitive function such as searching for information. BS. See base station. CDMA. See code division multiple access. CDPD. See cellular digital packet data. cellular digital packet data (CDPD). A standard for transmitting data over a cellular network that places the data in digital electronic envelopes and sends it at high speed through underused radio channels or during pauses in cellular phone conversations. Cellular Telecommunications Industry Association (CTIA). A trade group that represents cellular, PCS, and enhanced specialized mobile radio carriers. Challenge Handshake Authentication Protocol (CHAP). An authentication protocol that protects against eavesdropping by encrypting the user name and password. See also Password Authentication Protocol.
483
CHAP. See Challenge Handshake Authentication Protocol. CLNP. See Connectionless Network Protocol. CMRS. See Commercial Mobile Radio Services spectrum. code division multiple access (CDMA). A form of multiplexing where the transmitter encodes the signal using a pseudo-random sequence, which the receiver also knows and can use to decode the received signal. Each different random sequence corresponds to a different communication channel. Commercial Mobile Radio Services spectrum (CMRS). An organization that provides cellular, PCs, mobile radio, paging, and other wireless services. Connectionless Network Protocol (CLNP). An OSI protocol for the delivery of data. It uses datagrams (packets) that include address information for routing network messages. Used in local area networks (LANs) rather than wide area networks (WANs). copper chip technology. A technology in which chips exploit the capabilities of copper circuitry, which passes electrical currents more easily than aluminum. Aluminum has been the traditional material used in the wires that connect the switches, or transistors, in silicon chips. Copper chip technology produces smaller, faster chips that have enormous capacity for holding and transmitting information. CTIA. See Cellular Telecommunications Industry Association. DARPA. See Defense Advanced Research Projects Agency. data communication equipment (DCE). A device that establishes, maintains and terminates a session on a network. It may also convert signals for transmission. It is typically the modem. Data Encryption Standard (DES). A cryptographic algorithm designed to encrypt and decrypt data using a private key. DataTAC. A technology for radio packet-data networks that was developed by Motorola. data terminal equipment (DTE). A communications device that is the source or destination of signals on a network. It is typically a terminal or computer. DB2. A family of IBM licensed programs for relational database management. DCE. See data communication equipment. DCT. See Digital Cordless Telephone.
DECT. See Digital European Cordless Telecommunications. Defense Advanced Research Projects Agency (DARPA). The United States Department of Defense agency responsible for creating ARPANET, a large TCP/IP network. This agency was formerly called the Advanced Research Projects Agency (ARPA). DES. See Data Encryption Standard. DHCP. See Dynamic Host Configuration Protocol. Digital Cordless Telephone (DCT). A cordless telephone that uses digital transmission technology to provide more security and better sound quality than a traditional analog cordless telephone. Digital European Cordless Telecommunications (DECT). A digital wireless telephone technology that is expected to make cordless phones much more common in both businesses and homes. Digital Subscriber Line (DSL). A technology that dramatically increases the digital capacity of ordinary telephone lines in the home or office. DNS. See Domain Name System. Domain Name System (DNS). The distributed database system used to map domain names to IP addresses. DSL. See Digital Subscriber Line. DTE. See data terminal equipment. Dynamic Host Configuration Protocol (DHCP). A communications protocol that is used to centrally manage configuration information. For example, DHCP automatically assigns IP addresses to computers in a network. The Dynamic Host Configuration Protocol is defined by the Internet Engineering Task Force (IETF). EDGE. See Enhanced Data GSM Environment. electronic serial number (ESN). Unique identification number embedded in a wireless phone by the manufacturer to prevent fraud. ESN differs from the mobile identification number, which is the wireless carriers identifier for a phone in the network. embedded system. A computer system that is a component of a larger machine or system. Embedded systems can respond to events in real time. Hosts of embedded systems include watches, household appliances, cars, and cellular phones. Enhanced Data GSM Environment (EDGE). A faster version of the GSM wireless service that is designed to deliver data at rates up to 384 Kbps and enable the delivery of multimedia and other broadband applications to mobile phone and computer users. The EDGE standard is built on the existing GSM standard,
484
using the same time-division multiple access (TDMA) frame structure and existing cell arrangements. EPOC. An operating system designed for mobile devices. ESN. See electronic serial number. E-TDMA. See extended time division multiple access. extended time division multiple access (E-TDMA). See also Time Division Multiple Access. Extensible Markup Language (XML). A standard metalanguage for defining markup languages that is based on Standard Generalized Markup Language (SGML). XML simplifies the process of authoring and managing structured information and transmitting and sharing structured information across diverse computing systems. File Transfer Protocol (FTP). In TCP/IP, an application protocol used for transferring files to and from host computers. fixed wireless data. Wireless service to a fixed location through antennas larger than those in mobile or portable setups. The fastest data throughputs--up to T-1 speed--are available over fixed wireless networks. FTP. See File Transfer Protocol. full user mobility. Wireless classification under which users can access data while in motion, for example, in a vehicle. gateway. A ground-based link to a mobile satellite service network. General Packet Radio Service (GPRS). A packet data communications system that allows information to be sent and received across a mobile telephone network. GPRS permits faster Internet access and improved mobile technology through continuous connectivity. GEOS. See geo-stationary Earth orbit systems. geo-stationary Earth orbit systems (GEOS). Communications system with satellites in geosynchronous orbits 22,300 miles above Earth. global mobile personal communications services (GMPCS). Future mobile satellite systems that will provide global wireless phone service. global positioning system (GPS). A satellite constellation that provides highly accurate position, velocity, and time navigation information to users. Global System for Mobile Communications (GSM). A standard for digital cellular telephone systems that originated in Europe and is now used in countries across the globe. GSM networks support voice, data, text, and facsimile transmissions.
GMPCS. See global mobile personal communications services. GPRS. See General Packet Radio Service. GPS. See global positioning system. GSM. See Global System for Mobile Communications. Handheld Device Markup Language (HDML). A specialized version of HTML designed to enable wireless pagers, cell phones and other handheld devices to obtain information from Web pages. HDML. See Handheld Device Markup Language. HLR. See home location register. home location register (HLR). A database in a cellular system that contains all the subscribers within the providers home service area. HTTP. See Hypertext Transfer Protocol. HTTP-NG. See HyperText Transfer Protocol-Next Generation. Hypertext Transfer Protocol (HTTP). An Internet protocol that is used to transfer and display hypertext and XML documents on the Web. HyperText Transfer Protocol-Next Generation (HTTP-NG). A replacement for HTTP 1.0, HTTP-NG maintains HTTP 1.0s simplicity while adding important features such as security and authentication. ICMP. See Internet Control Message Protocol. iDEN. See integrated digital enhanced network. independent software vendor (ISV). A non-IBM software vendor. infrared technology (IR). A technology that uses infrared radiation for wireless transmission between computer devices, as well as many handheld remotes for TVs and video and stereo equipment. Infrared transmission requires an unobstructed line of sight between transmitter and receiver. integrated digital enhanced network (iDEN). A Motorola-enhanced mobile radio network technology that integrates two-way radio, telephone, text messaging, and data transmission into a single network. Integrated Services Digital Network (ISDN). An international communications standard for sending voice, video, and data over digital telephone lines. Internet Control Message Protocol (ICMP). An Internet protocol that is used by a gateway to communicate with a source host, for example, to report an error in a datagram.
Glossary
485
Internet screenphone. A telephone-like appliance with a built-in display screen that may be used to call up Internet sites quickly. Internet service provider (ISP). An organization that provides access to the Internet. in-vehicle information system (IVIS). A system that enables intelligent vehicles to provide the functionality to increase the productivity and safety of the driver. IR. See infrared technology. IS-95. Spread spectrum air interface technology used in some digital cellular and personal communications services, and other wireless networks. ISDN. See Integrated Services Digital Network. ISP. See Internet service provider. ISV. See independent software vendor. ITU-2000. The new third generation global standard for mobile telecommunications, coined by the International Telecommunications Union. Also known as IMT-2000. IVIS. See in-vehicle information system. Java. An object-oriented programming language for portable interpretive code that supports interaction among remote objects. Java was developed and specified by Sun Microsystems, Incorporated. Jini. A trademark for technology licensed by Sun Microsystems. Jini is a Java-based connection technology. It allows Jini-compatible devices such as printers to connect transparently to a network and to interact. kiosk services. Solutions for business center service providers such as hotel chains, airports and office service franchises that offer document services (storing, reproduction and distribution) through shared access devices in public spaces. LAN. See local area network. LDAP. See Lightweight Directory Access Protocol. LEC. See local exchange carrier. LEO. See low earth orbit. Lightweight Directory Access Protocol (LDAP). An open protocol that uses TCP/IP to provide access to directories that support an X.500 model and that does not incur the resource requirements of the more complex X.500 Directory Access Protocol (DAP). For example, LDAP can be used to locate people, organizations, and other resources in an Internet or intranet directory.
Linux. An open source implementation of the UNIX system. LMDS. See Local Multipoint Distribution Service. LNP. See local number portability. local area network (LAN). A network that connects several devices into a limited area (such as a single building or campus) and that can be connected to a larger network. local exchange carrier (LEC). An authorized carrier that has been commissioned to provide local voice-level telecommunications services within a predetermined area. Local Multipoint Distribution Service (LMDS). A system for broadband microwave wireless transmission direct from a local antenna to homes and businesses within a line-of-sight radius; a solution to the so-called last-mile technology problem of economically bringing high-bandwidth services to users. LMDS is an alternative to installing optical fiber all the way to the user or to adapting cable TV for broadband Internet service. local number portability (LNP). Subscribers ability to switch local or wireless carriers and still retain the same phone number, as is possible now with long-distance carriers. low earth orbit (LEO). An orbit between 700 and 2,000 kilometers above the Earth that is used by mobile communications satellites. MAN. See metropolitan area network. MDSP. See Mobile Data Synchronization Protocol. MDSS. See Mobile Data Synchronization Service. MEMS. See Micro-Electric Mechanical System. metropolitan area network (MAN). A network formed by the interconnection of two or more networks which may operate at higher speed than those networks, may cross administrative boundaries, and may use multiple access methods. See also wide area network. microbrowser. A Web browser designed for small display screens on smart phones and other handheld wireless devices. Micro-Electric Mechanical System (MEMS). A technology that embeds mechanical devices such as fluid sensors, mirrors, actuators, pressure and temperature sensors, vibration sensors and valves in semiconductor chips. middleware. Software that connects two separate applications and passes data between them. For example, a type of middleware can link a database
486
system to a Web server. This allows users to request information from the database using forms displayed on a Web browser, and it enables the server to return dynamic Web pages based on the users request and profile. MNCRS. See Mobile Network Computing Reference Specification. Mobile Data Synchronization Protocol (MDSP). A protocol that defines the form of an XML document for data exchange. Insertions, modifications, and deletions of items in a data store can be described in MDSP documents. Mobile Data Synchronization Service (MDSS). The ability to synchronize data on client devices with data stored in an enterprise database. A common protocol has been developed that will eventually allow a variety of clients to synchronize with a variety of databases. Mobile Network Computing Reference Specification (MNCRS). A specification that defines a set of standards for mobile Java devices. mobile satellite service (MSS). A radiocommunication service between mobile earth stations and one or more space stations, or between space stations used by this service. mobitex. An open standard for sharing information among mobile devices using radio transmission. MQLite. A lightweight message queuing facility that provides a subset of MQ Series functionality. MQLite transmits Mobile Data Synchronization Protocol (MDSP) documents between the client and the mid-tier server. MSS. See mobile satellite service. multi-modal. Pertaining to a system that operates using multiple interfaces (e.g. both text and speech). Multi-modal paradigm user interfaces (MMUIs) will be used in next-generation devices to allow interaction via voice, touch and keyboard input. NAMPS. See Narrowband Advanced Mobile Phone Service. Narrowband Advanced Mobile Phone Service (NAMPS). A standard that combines cellular voice processing with digital signaling, increasing the capacity of Advanced Mobile Phone Service (AMPS) systems and adding functionality. National Telecommunications and Information Administration (NTIA). The United States governments authority on spectrum management. Network News Transfer Protocol (NNTP). A protocol that is used to post messages in, distribute messages to,
and retrieve messages from newsgroups and to transfer articles between news servers. NNTP. See Network News Transfer Protocol. NTIA. See National Telecommunications and Information Administration. ODBC. See Open Database Connectivity. Open Database Connectivity (ODBC). A standard application programming interface (API) for accessing data in both relational and nonrelational database management systems. Using this API, database applications can access data stored in database management systems on a variety of computers even if each database management system uses a different data storage format and programming interface. Open Service Gateway (OSG). A specification that is being defined by the Open Service Gateway initiative (OSGi), a consortium of more than 20 companies, including IBM. The OSG specification will outline open standards for the management of voice, data and multimedia wireless and wired networks.. open systems interconnection (OSI). The interconnection of open systems in accordance with standards of the International Organization for Standardization (ISO) for the exchange of information. OSG. See Open Service Gateway. OSI. See open systems interconnection. Palm OS platform. An operating system for handheld devices, designed for information management. PAN. See personal area network. PAP. See Password Authentication Protocol. Password Authentication Protocol (PAP). A type of authentication where the user name and password are transmitted in an unencrypted form. This is a more basic form of authentication than Challenge Handshake Authentication Protocol (CHAP). See also Challenge Handshake Authentication Protocol. PCMCIA. See Personal Computer Memory Card International Association. PCS. See personal communication system. PDA. See personal digital assistant. PDC. See personal digital cellular. personal area network (PAN). Experimental IBM technology that allows information devices to communicate discreetly and synchronize with one another when in close proximity, using the human body as an antenna.
Glossary
487
personal communication system (PCS). A Federal Communications Commission classification for digital wireless communication systems based on the same principles as cellular systems, but usually operating in a different frequency range and with smaller cell sizes than cellular systems. Personal Computer Memory Card International Association (PCMCIA). An international standards body and trade association founded in 1989 to establish a standard for connecting peripherals to portable computers. personal digital assistant (PDA). A handheld device that is used for personal organization tasks (such as calendaring, note-taking, and recording telephone and fax numbers), and networking functions such as email and synchronization. personal digital cellular (PDC). Technology used in digital cellular telephone communication in Japan. Personal Handy Phone System (PHS). Personal communications system used in Japan as both a digital cordless phone in homes and offices and as a cellular phone outdoors. PHS. See Personal Handy Phone System. Pocket PC. A handheld Windows-based computer that runs the Pocket PC operating system (formerly Windows CE). Point-to-Point Protocol (PPP). A data-link protocol for communication between two computers that use a serial interface, typically a personal computer connected by telephone line to a server. POP3. See Post Office Protocol Version 3. portable wireless data. A classification of networks that access the Internet through a laptop or palmtop computer and a small wireless modem. Post Office Protocol Version 3 (POP3). A mail protocol that allows clients running it to retrieve mail from a host mail server also running the protocol. PPP. See Point-to-Point Protocol. radio frequency (RF). An alternating current that generates an electromagnetic field when applied to an antenna. The generated electromagnetic field is suitable for wireless broadcasting and communications. RAM. See random access memory. random access memory (RAM). Computer memory in which any storage location can be accessed directly. RC5. An encryption technology for use in wireless clients and servers. RF. See radio frequency.
Scalable Vector Graphics (SVG). A vector graphics language written in XML. SDK. See software development kit. SDSL. See symmetric digital subscriber line. Secure Sockets Layer. A security protocol that provides communication privacy. With SSL, client/server applications can communicate in a way that is designed to prevent eavesdropping, tampering, and message forgery. set-top box. A single or multiple tuner device that enables the display of cable or satellite signals on a television. SGML. See Standard Generalized Markup Language. Short Message Service (SMS). A service that is used to transmit text to and from a mobile phone. SiGe. See silicon-germanium technology. Signaling System 7 (SS7). The international high-speed signaling backbone used for the public-switched telephone network. silicon-germanium technology (SiGe). A technology that makes possible the design of complex chips that integrate the functions of a cellular telephone, an e-mailbox and an Internet browser into a hand-held information device with rapid data transfer capability. Silicon-On-Insulator (SOI). A microchip technology that provides faster computer chips that also require less power, a key requirement for extending the life of pervasive devices. SIM. See subscriber identity module. Simple Mail Transfer Protocol. An Internet application protocol for transferring mail among users of the Internet. SMTP specifies the mail exchange sequences and message format. It assumes that the Transmission Control Protocol (TCP) is the underlying protocol. smart phone. A phone that enables connections to the Internet or Lotus Notes network so that users can access e-mail, faxes, voice mail, Web pages and other files. It can also connect to a speech recognition server by phone so that a user can dictate notes or responses and then see the recognized text as it is returned from the server. SMF. See System Management Facility. SMS. See Short Message Service. software development kit (SDK). A set of tools, APIs, and documentation to assist with the development of software in a specific computer language or for a particular operating environment.
488
SOI. See Silicon-On-Insulator. SQL. See Structured Query Language. SS7. See Signaling System 7. Standard Generalized Markup Language (SGML). A standard metalanguage for defining markup languages that is based on the ISO 8879 standard. SGML focuses on structuring information rather than presenting information; it separates the structure and content from the presentation. It also facilitates the interchange of documents across an electronic medium. Structured Query Language (SQL). A standardized language for defining and manipulating data in a relational database. subscriber identity module (SIM). A smart card inserted into GSM phones that contains the users telephone account information, and that allows the user to use a borrowed or rented GSM phone as if it were his or her own. SVG. See Scalable Vector Graphics. symmetric digital subscriber line (SDSL). A technology that uses digital pulses to send data over copper telephone lines. SDSL cannot operate simultaneously with voice connections over the same wires. System Management Facility (SMF). A z/OS facility that collects and records a variety of system and job-related information. Examples of information collected by SMF are statistics, accounting information, and performance data. TACS. See Total Access Communication System. TCP/IP. See Transmission Control Protocol/Internet Protocol. TDMA. See Time Division Multiple Access. third generation (3G). The next generation of wireless technology. Multimedia 3G networks transmit wireless data up to 2 megabits per second, making possible the integration of voice, data and video. Time Division Multiple Access (TDMA). A digital communication technology used by some carriers to provide PCS service. TDMA is a multiplexing technology used with satellites and cell phones, in which multiple channels of data are interleaved (each signal is assigned to a different time interval, and the signals are transmitted by taking turns). See also extended time division multiple access. Time-Modulated Ultra-Wide Band (TM-UWB). New wireless technology that greatly broadens bandwidth. Developed by Time Domain, TM-UWB uses short, coded pulses transmitted over a wide range of frequencies, making it useful for a broad range of
applications from networking to through-the-wall radar and secure communications. TM-UWB. See Time-Modulated Ultra-Wide Band. Total Access Communication System (TACS). An analog cellular phone system deployed mostly in Europe. It was modeled after the AMPS system in the U.S. transcoding. The operation of changing data from one format to another, such as XML to HTML, so that the output will be displayed in a manner appropriate to the device. Transmission Control Protocol/Internet Protocol (TCP/IP). An industry-standard nonproprietary set of communications protocols that provide reliable end-to-end connections between applications over interconnected networks of different types. Two-Party Key Distribution Protocol (2PKDP). A security protocol that combines bidirectional authentication with key distribution using a minimal number of messages. UDP. See User Datagram Protocol. UMTS. See Universal Mobile Telecommunications System. Universal Mobile Telecommunications System (UMTS). The third generation mobile telecommunications standard, defined by the ITU, that increases transmission speed to 2 Mbps per mobile user and establishes a global roaming standard. User Datagram Protocol (UDP). An Internet protocol that provides unreliable, connectionless datagram service. It enables an application program on one machine or process to send a datagram to an application program on another machine or process. virtual private network (VPN). An extension of a companys intranet over the existing framework of either a public or private network. A VPN ensures that the data that is sent between the two endpoints of its connection remains secure. Voice over Internet Protocol (VoIP). Sending telephony voice over Internet Protocol (IP) data connections rather than existing dedicated voice networks, switching and transmission equipment. VoIP. See Voice over Internet Protocol. VPN. See virtual private network. WAN. See wide area network. WAP. See Wireless Application Protocol. WBXML. See Wireless Binary XML.
Glossary
489
W-CDMA. See wideband code division multiple access. WDF. See Wireless Data Forum. Web pad. A handheld, touch-screen wireless device for Web browsing. wide area network (WAN). A network that provides communication services between devices in a geographic area larger than that served by a local area network (LAN) or a metropolitan area network (MAN). See also metropolitan area network. wideband code division multiple access (W-CDMA). A CDMA channel that is four times wider than the channels that are typically used in 2G networks in North America. It supports very high-speed multimedia services such as full-motion video. It uses one 5 MHz channel for both voice and data, offering data speeds up to 2 Mbps. Wideband CDMA is known as UMTS in Europe. WiFi. See Wireless Fidelity. WIN. See wireless intelligent network. WinCE. See Windows CE. Windows CE (WinCE). A version of the Windows operating system designed for handheld devices. Version 3 is often known as Pocket PC. Wireless Application Protocol (WAP). An open industry standard for mobile Internet access that allows mobile users with wireless devices to easily and instantly access and interact with information and services. Wireless Binary XML (WBXML). A succinct encoding of XML defined by the WAP Forum. MDSP documents are encoded into WBXML and the WBXML byte stream is sent by enqueuing it on an MQLite queue as a single message object. Wireless Data Forum (WDF). An industry group dedicated to promoting wireless data. Wireless Fidelity (WiFi). A high-frequency wireless local area network. WiFi technology is rapidly gaining acceptance in many companies as an alternative to a wired LAN. It can also be installed for a home network. wireless intelligent network (WIN). Wireless switched network that allows carriers to provide enhanced and customized services for mobile phones. wireless LAN. A network that uses wireless transmissions such as radio or infrared to provide communication services between devices. Wireless Markup Language (WML). A markup language based on XML that is used to present content
and user interfaces for wireless devices such as cellular phones, pagers, and personal digital assistants. wireless network. See wireless LAN. Wireless Session Protocol (WSP). A sandwich layer that links the Wireless Application Environment (WAE) to two session services: one connection-oriented, operating above the Wireless Transaction Protocol, the other connectionless, operating above the Wireless Datagram Protocol. WML. See Wireless Markup Language. WSP. See Wireless Session Protocol. XML. See Extensible Markup Language.
490
Index A
accessibility features 7 addNode command 96, 99 administrative console configuring engines 93 filtering 121 icons 120 login settings 119 navigation tree 120 starting 117 starting a server 125 stopping 117 stopping a server 125 taskbar 120 using 117 workspace 120 ASR settings languages installed 93 number of engines 93 preinstalling grammars 93
G
Genesys support 112 German phonetic alphabet 22 getParam sample 127 GVP installation prerequisites 88
P
planning voice system 43 planning resources ASR engines 47 LAN 48 TTS engines 47 post-installation 85 pronunciations 20 putParam sample 128
H
hardware requirements HTTP server starting 142 stopping 142 48
R
Red Hat prerequisites 60 removeNode command 99 response file 83 restoreVoiceServerConfig sample backup_file parameter 134 global security 134 help parameter 134
I
IBM Support collecting information 458 collector tool 458 IBM TTS Connector configuration parameters 113 IBMTextToSpeech.ini file 113 identifiers 452 Information Center 13 information conventions 15 installation errors 475 installation prerequisites 60 installation resources 83 installation tips 83 installation verification test 83 troubleshooting 476 installing fix pack 79 IP configurations 60
B
backupVoiceServerConfig sample backup_file parameter 132 global security 132 help parameter 132
S
sample scripts backupVoiceServerConfig 132 getParam 127 putParam 128 restoreVoiceServerConfig 134 startVoiceServer 129 statusVoiceServer 131 stopVoiceServer 130 scripting 95 samples 126 wsadmin tool 126 security 9, 125 Simplified Chinese phonetic alphabet software requirements 48 speech applications 18 speech technology business advantages 1 overview 2 standards for developing grammars 17 TTS 17 startServer 129 startVoiceServer sample global security 129 help parameter 129 statusVoiceServer sample global security 131 help parameter 131 stopVoiceServer sample global security 130 help parameter 130 wvsonly parameter 130
C
Canadian French phonetic alphabet collector tool 458 command line tools 95 configuration parameters 105 Configuration Service MBean 102 configuration settings 122 console extensions 81 creating custom scripts 101 21
J
Japanese phonetic alphabet 24
27
D
deployment manager 80, 81
L
Latin American Spanish phonetic alphabet 26 LaunchPad 61 Load Balancer 82 log messages 154
E
Edge Component 82 environments multiple machines 54 single machine 50
M
message IDs 154 messages 154 MIME mapping 105
F
First Steps 85 firststeps command 85
N
Network Deployment uninstalling 90 notices 481 53, 54, 80, 81
T
testing grammars 18
491
topologies multiple machines 53 tracing 453 trademarks 481 troubleshooting installation 475 installation verification test 476 messages 153 testing 476 troubleshooting tips administrative 461 ASR 467 gateways 474 IBM TTS Connector 473 installation 461 management 477 runtime 477 security 462 TTS 463 voiceTest 474 TTS settings default voice 94 number of engines 94 preloading grammars 94 voices installed 94 tuning configuration settings 145 AIX 147 Cisco gateway 150 for Avaya 145 for Cisco 145 for Genesys 145 Linux 148 SUSE 149 WebSphere Application Server 146 Windows 149 speech resources 19
WebSphere Voice Server access 8 features 6 installing 57, 59, 68 messages 153 overview 4 package 58 previous versions 11 resources for learning 479 restarting server processes 135 scalability 9 silent installation 76 tools 10 troubleshooting tools 9 uninstalling 88 WebSphere Voice Toolkit 10 WVSAdminConfig script 101
U
UK English phonetic alphabet 38 uninstalling administrative console extensions 90 US English phonetic alphabet 39 user dictionary 20
V
voice server adding 96 removing 99 Voice Servers panel configuration problems 122 runtime messages 122 server status 122 Voice Toolkit 18 VoiceXML 17
W
WebSphere Application Server manually uninstalling 90 uninstalling 90 WebSphere family 12 WebSphere Voice Response support
112
492
Readers Comments Wed Like to Hear from You

Voice Server WebSphere Voice Server for Multiplatforms Version 5.1.3 Overall, how satisfied are you with the information in this book? Very Satisfied Overall satisfaction h Satisfied h Neutral h Dissatisfied h Very Dissatisfied h
How satisfied are you that the information in this book is: Very Satisfied Accurate Complete Easy to find Easy to understand Well organized Applicable to your tasks h h h h h h Satisfied h h h h h h Neutral h h h h h h Dissatisfied h h h h h h Very Dissatisfied h h h h h h
Please tell us how we can improve this book:
Thank you for your responses. May we contact you?
h Yes
h No
When you send comments to IBM, you grant IBM a nonexclusive right to use or distribute your comments in any way it believes appropriate without incurring any obligation to you. IBM or any other organizations will only use the personal information that you supply to contact you about the issues that you state on this form.
Name Company or Organization Phone No.
Address
___________________________________________________________________________________________________
Readers Comments Wed Like to Hear from You
Cut or Fold Along Line
Fold and _ _ _ _ _ _ _ _ _ _Fold and_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _Please _ _ _ _ _ staple _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Tape _ _ _ _ _ _ _ _ Tape _ _ _ _ do not _ _ _ _ NO POSTAGE NECESSARY IF MAILED IN THE UNITED STATES
BUSINESS REPLY MAIL

FIRST-CLASS MAIL PERMIT NO. 40 ARMONK, NEW YORK POSTAGE WILL BE PAID BY ADDRESSEE
IBM Corporation Information Development Department MMOA P.O. Box 12195 Research Triangle Park, NC 27709-9990
_________________________________________________________________________________________ Please do not staple Fold and Tape Fold and Tape
Cut or Fold Along Line

WVS For Multi Platforms v5.1 Product Documentation

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

WVS For Multi Platforms v5.1 Product Documentation

Uploaded by

Copyright:

Available Formats

WebSphere Voice Server

WebSphere Voice Server for Multiplatforms

Chapter 6. Administering . . . . . . 117

Chapter 8. Troubleshooting . . . . . 153

Glossary . . . . . . . . . . . . . 483 Index . . . . . . . . . . . . . . . 491

WebSphere Voice Server for Multiplatforms

Chapter 1. Product overview

Copyright IBM Corp. 2005, 2006

Brief introduction to speech

How does ASR work?

WebSphere Voice Server for Multiplatforms

How does TTS work?

What does WebSphere Voice Server do?

Speech Recognition Grammar Specification Version 1.0

WebSphere Voice Server for Multiplatforms

WebSphere Voice Server for Multiplatforms

Multiple language support

Chapter 1. Product overview

Figure 1. The MRCP environment

WebSphere Voice Server for Multiplatforms

Administrative ease of use

Chapter 1. Product overview

Compatibility with IVRs and gateways

WebSphere Voice Server for Multiplatforms

Rapid problem determination

Product GUIs and tools

Chapter 1. Product overview

Tools for installing

Tools for developing applications

Tools for troubleshooting

WebSphere Voice Server for Multiplatforms

Relationship to other versions of WebSphere Voice Server products

Chapter 1. Product overview

WebSphere platform and related software

Foundation and tools

Reach and user experience products

Business integration products

WebSphere Voice Server for Multiplatforms

About this information

Information center change history

Chapter 1. Product overview

Date December 12, 2005

June 14, 2005

March 11, 2005

January 23, 2005

WebSphere Voice Server for Multiplatforms

Date November 23, 2004

July 16, 2004

Chapter 1. Product overview

WebSphere Voice Server for Multiplatforms

Standards for developing

Copyright IBM Corp. 2005, 2006

Developing a speech application

WebSphere Voice Server for Multiplatforms

Testing Japanese grammars

Tuning speech resources for effective applications

Preparing a lexicon of ASR pronunciations

Canadian French phonetic pronunciation

WebSphere Voice Server for Multiplatforms

German phonetic pronunciation

WebSphere Voice Server for Multiplatforms

Japanese phonetic pronunciation