From: Dynamically localizing multiple speakers based on the time-frequency domain
Distance | 1 m | 2 m | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
RT60 | 0.160 s | 0.360 s | 0.610 s | 0.160 s | 0.360 s | 0.610 s | ||||||
Measure | MAE | Acc. | MAE | Acc. | MAE | Acc. | MAE | Acc. | MAE | Acc. | MAE | Acc. |
MUSIC | 18.7 | 57.6 | 19.2 | 53.2 | 21.9 | 42.9 | 18.4 | 54.1 | 26.1 | 35.8 | 25.4 | 32.2 |
SRP-PHAT | 9.0 | 39.0 | 13.9 | 39.4 | 18.6 | 29.9 | 9.7 | 36.0 | 16.5 | 24.7 | 27.7 | 21.3 |
CMS-DOA | 1.6 | 76.3 | 7.3 | 75.2 | 8.4 | 71.9 | 5.1 | 79.5 | 9.7 | 60.1 | 17.5 | 40.0 |
TF-DOAnet | 1.3 | 97.5 | 3.5 | 83.5 | 0.9 | 98.3 | 5.0 | 89.5 | 1.7 | 95.7 | 4.8 | 84.2 |