Speech Separation On Lrs2
評価指標
PESQ
SDRi
SI-SNRi
STOI
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | PESQ | SDRi | SI-SNRi | STOI |
|---|---|---|---|---|
| tdfnet-an-efficient-audio-visual-speech | 3.16 | 15.2 | 15.0 | 0.938 |
| rtfs-net-recurrent-time-frequency-modelling | - | 15.1 | 14.9 | - |
| tdfnet-an-efficient-audio-visual-speech | 3.10 | 13.7 | 13.6 | 0.931 |
| scanet-a-self-and-cross-attention-network-for | - | 16.6 | 16.4 | - |
| an-audio-visual-speech-separation-model | - | - | 14.3 | - |
| rtfs-net-recurrent-time-frequency-modelling | - | 14.8 | 14.6 | - |
| tdfnet-an-efficient-audio-visual-speech | 3.21 | 15.9 | 15.8 | 0.949 |
| rtfs-net-recurrent-time-frequency-modelling | - | 14.3 | 14.1 | - |