Skip to main content

Table 8 Performance improvement for ULAs with different number of microphones when applied in rooms with different reverberation times

From: A multichannel learning-based approach for sound source separation in reverberant environments

Channel

Model

ΔSI-SNR (dB)

ΔPESQ

ΔSTOI

0.16 s

0.36 s

0.61 s

0.9 s

Avg.

0.16 s

0.36 s

0.61 s

0.9 s

Avg

Avg.

2-ch

BF-net + U-net + DM

9.43

10.60

12.04

11.94

11.02

0.55

0.48

0.42

0.32

0.44

0.15

4-ch

11.31

12.52

14.57

14.69

13.29

0.91

0.74

0.59

0.49

0.68

0.20

6-ch

11.37

12.51

14.61

14.72

13.32

0.92

0.76

0.60

0.47

0.69

0.20

Score of unprocessed signals

− 6.10

− 11.94

− 16.24

− 18.50

− 13.25

1.82

1.60

1.44

1.38

1.56

0.59

  1. The best scores for the respective subset are boldfaced