You are on page 1of 2

Dcomposition de signal unidimentionnel(signal vocal):

y=wavread('test_07b.wav');
[c,l]=dwt(y,'Haar');
figure(1)
subplot(2,2,1);plot(y);
title('signal vocal original');
subplot(2,2,2);plot(c);
title('signal d approximation')
subplot(2,2,3);plot(l);
title('signal de detail')

signal vocal original

signal d aproximation

0.5

1
0.5

0
-0.5

-0.5

2000

4000

-1

6000

1000

2000

3000

signal de detail
0.2
0.1
0
-0.1
-0.2

[c,l]=wavedec(y,3,'Haar');
l3=detcoef(c,l,3);
l2=detcoef(c,l,2);
l1=detcoef(c,l,1);
figure(2)
subplot(2,2,1);plot(l3);
subplot(2,2,2);plot(l2);
subplot(2,2,3);plot(l1);

1000

2000

3000

detail vertical

detail horizontal

0.5

0.6
0.4

0.2
0

-0.5

200

400

600

800

-0.2

500

1000

detail orthogonal
0.2
0.1
0
-0.1
-0.2

1000

2000

3000

Reconstruction du signal vocal:


0.5

yr=waverec(c,l,'Haar'
figure(3)
plot(yr)

0.4
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
-0.4
-0.5

1000

2000

3000

4000

5000

6000

1500

lalgorithme de Daubechies

signal vocal original

signal d aproximation

0.5

y=wavread('test_07b.wav');
[c,l]=dwt(y,'db10');
figure(1)
subplot(2,2,1);plot(y);
title('signal vocal original');
subplot(2,2,2);plot(c);
title('signal d approximation')
subplot(2,2,3);plot(l);

1
0.5

0
-0.5

-0.5

2000

4000

6000

1000

2000

3000

0.2
0.1
0
-0.1
-0.2

-1

1000

2000

3000

You might also like