Table 2 SDRi/SI-SNRi (dB) performance of Conv-TasNet and DPCCN on Libri2Mix, WHAMR!, and Aishell2Mix test sets. Systems are all trained on Libri2Mix
From: Heterogeneous separation consistency training for adaptation of unsupervised speech separation
System | SDRi/SI-SNRi (dB) | ||
|---|---|---|---|
Libri2Mix | WHAMR! | Aishell2Mix | |
Conv-TasNet | 12.41/11.98 | 6.83/6.45 | 2.57/2.08 |
DPCCN | 13.48/13.04 | 8.99/8.50 | 5.78/5.09 |