From: Localization based stereo speech source separation using probabilistic time-frequency masking and deep neural networks
Room
Type
ITDG (ms)
DRR (dB)
T60 (s)
A
Medium office
8.73
6.09
0.32
B
Small class room
9.66
5.31
0.47
C
Large lecture theatre
11.9
8.82
0.68
D
Large seminar room
21.6
6.12
0.89