Audio-visual speech separation on LRS2 2Mix
15.9SDRiTDFNet-large
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| TDFNet-largeParams (M)=6.5, MACs (G)=47.22024.01 | 15.9 | 15.8 | 3.21 | 0.949 | |
| TDFNet (MHSA + Shared)Params (M)=4.2, MACs (G)=38.62024.01 | 15.2 | 15 | 3.16 | 0.938 | |
| CTCNetParams (M)=7.1, MACs (G)=167.22024.01 | 14.6 | 14.3 | 3.08 | 0.931 | |
| TDFNet-smallParams (M)=5.8, MACs (G)=152024.01 | 13.7 | 13.6 | 3.1 | 0.931 | |
| CaffNet-C*Notes=Variant of CaffNet-C2024.01 | 12.5 | — | 1.15 | 0.89 | |
| VisualVoiceParams (M)=77.82024.01 | 11.8 | 11.5 | 3 | — | |
| Thanh-Dat2024.01 | 11.6 | — | 3.1 | — | |
| Conv-TasNetParams (M)=5.62024.01 | 10.7 | 10.3 | — | — | |
| A-FRCNNParams (M)=6.32024.01 | 10.1 | 9.4 | — | — | |
| CaffNet-C2024.01 | 10 | — | 0.94 | 0.88 | |
| SuDORM-RFParams (M)=2.72024.01 | 9.5 | 9.1 | — | — | |
| uPITParams (M)=92.72024.01 | 4.8 | 3.6 | — | — |