Table 2 Speech separation accuracy of ODAN in separating one-, two-, and three-speaker mixtures (WSJ0-mix2 and WSJ0-mix3 datasets).

Speech separation accuracy of ODAN in separating one-, two-, and three-speaker mixtures (WSJ0-mix2 and WSJ0-mix3 datasets).. The ODAN was trained on both the WSJ0-mix2 and WSJ0-mix3 datasets and used in all cases.

Number of
speakers
CausalSI-SNRi (dB)SDRi (dB)PESQESTOI
3Yes7.07.52.080.56
2Yes8.99.32.630.76
1YesSI-SNR (dB)
24.4
SDR (dB)
25.0
4.140.98