### TABLE I GIBBS SAMPLER FOR SOURCE SEPARATION OF STUDENT t SOURCES.

2005

Cited by 6

### TABLE I SOURCE SEPARATION PERFORMANCE ON TWO-SOURCE MIXTURES.

### TABLE I EXPERIMENTS ON SOURCE SEPARATION FOR SEVERAL DISTRIBUTIONS OF SOURCES. ON PERFECT SEPARATION, THE RESULTING MATRIX M = AB SHOULD BE THE IDENTITY.

### Table 4 Geometry and Results of Source Separation Estimation

### Table 1. Log spectral distortion and segmental SNR for each of the 3 separated sources .

2004

"... In PAGE 3: ... In evaluating our post- lter, we use both the segmental SNR and the log spectral distortion (LSD), which is de ned as: LSD = 1 L L 1 X l=0 2 6 4 1 K K 1 X k=0 0 @20 log10 jX(k; l)j + ^ X(k; l) + 1 A 23 7 5 1 2 (13) where L is the number of frames and K is the number of frequency bins and is meant to prevent extreme values. Table1 compares the results for separation of each of the 3 original sources with the single-channel and the multi- channel Cohen post- lters, both described in [1]. The Co- hen post- lter is adapted to uses the other sources as refer- ence noise signals.... ..."

Cited by 7

### Table 1. Log spectral distortion and segmental SNR for each of the 3 separated sources.

2004

"... In PAGE 3: ... In evaluating our post- lter, we use both the segmental SNR and the log spectral distortion (LSD), which is de ned as: LSD = 1 L L 1 X l=0 2 6 4 1 K K 1 X k=0 0 @20 log10 jX(k; l)j + ^ X(k; l) + 1 A 23 7 5 1 2 (13) where L is the number of frames, K is the number of fre- quency bins and is meant to prevent extreme values. Table1 compares the results for separation of each of the 3 original sources with the single-channel and the multi- channel Cohen post- lters, both described in [1]. The Co- hen post- lter is adapted to uses the other sources as refer- ence noise signals.... ..."

Cited by 7

### TABLE I SIGNAL-TO-NOISE RATIO (SNR) FOR EACH OF THE THREE SEPARATED SOURCES.

2004

Cited by 14

### TABLE I SIGNAL-TO-NOISE RATIO (SNR) FOR EACH OF THE THREE SEPARATED SOURCES.

2004

Cited by 14