Audio-visual speech separation on TCD-TIMIT
1,090SDRVISUALVOICE
Evaluation results
| Method | Links | ||
|---|---|---|---|
| VISUALVOICEspeaker-dependency=speaker-independent, training_dataset=VoxCeleb22021.01 | 1,090 | 2.91 | |
| Ephrat et al.speaker-dependency=speaker-independent, training_dataset=AV-Speech2021.01 | 410 | 2.42 | |
| Gabbay et al.speaker-dependency=speaker-dependent2021.01 | 40 | 2.03 |