Speech Separation on LRS2
Metric: SI-SNRi (higher is better)
Metric: SI-SNRi (higher is better)
| # | Model↕ | SI-SNRi▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | IIANet | 16.4 | No | IIANet: An Intra- and Inter-Modality Attention N... | 2023-08-16 | Code |
| 2 | TDFNet-large | 15.8 | No | TDFNet: An Efficient Audio-Visual Speech Separat... | 2024-01-25 | Code |
| 3 | TDFNet (MHSA + Shared) | 15 | No | TDFNet: An Efficient Audio-Visual Speech Separat... | 2024-01-25 | Code |
| 4 | RTFS-Net-12 | 14.9 | No | RTFS-Net: Recurrent Time-Frequency Modelling for... | 2023-09-29 | Code |
| 5 | RTFS-Net-6 | 14.6 | No | RTFS-Net: Recurrent Time-Frequency Modelling for... | 2023-09-29 | Code |
| 6 | CTCNet | 14.3 | No | An Audio-Visual Speech Separation Model Inspired... | 2022-12-21 | Code |
| 7 | RTFS-Net-4 | 14.1 | No | RTFS-Net: Recurrent Time-Frequency Modelling for... | 2023-09-29 | Code |
| 8 | TDFNet-small | 13.6 | No | TDFNet: An Efficient Audio-Visual Speech Separat... | 2024-01-25 | Code |