Skip to main content
EURASIP Journal on Advances in Signal Processing

Table 2 Average input (mixed) SIRs in decibels for both corpora and for different M × N mixing matrices

From: Audio visual speech source separation via improved context dependent association model

Ch.i

Alpha digits

 

Poet verses

 

2×2

2×3

3×3

5×5

 

2×2

2×3

3×3

5×5

Ch. 1

0.6

−2.5

−3.2

−9.0

 

−0.7

−3.9

−4.5

−10.2

Ch. 2

−1.8

−4.8

−5.9

−8.6

 

−3.2

−6.1

−7.3

−9.9

Ch. 3

-

-

0.4

−8.8

 

-

-

−0.9

−10.1

Ch. 4

-

-

-

−8.2

 

-

-

-

−9.4

Ch. 5

-

-

-

−6.7

 

-

-

-

−8.0

  1. SIRs values are calculated with respect to s1 (SIR(xi|s1)) and they are averaged over all simulated input random matrices and all frames for each corpus.
Back to article page