TDFNet (MHSA + Shared)

Reported on 4 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Speech4 results

Speech SeparationonLRS2
PESQ· 2024-01-25
3.16
best: 3.21 (TDFNet-large)
TDFNet: An Efficient Audio-Visual Speech Separation Model with Top-down Fusion arXiv:2401.14185
Speech SeparationonLRS2
SDRi· 2024-01-25
15.2
best: 16.6 (IIANet)
TDFNet: An Efficient Audio-Visual Speech Separation Model with Top-down Fusion arXiv:2401.14185
Speech SeparationonLRS2
SI-SNRi· 2024-01-25
15
best: 16.4 (IIANet)
TDFNet: An Efficient Audio-Visual Speech Separation Model with Top-down Fusion arXiv:2401.14185
Speech SeparationonLRS2
STOI· 2024-01-25
0.938
best: 0.949 (TDFNet-large)
TDFNet: An Efficient Audio-Visual Speech Separation Model with Top-down Fusion arXiv:2401.14185