Speech Enhancement on Speech Enhancement (SE) Task (test)
2.155PESQRNDVoC-shared
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| RNDVoC-sharedSE Model Type=Discriminative, Training Steps=1M2026.03 | 2.155 | 81.87 | 3.292 | 3.695 | |
| BigVGANSE Model Type=Discriminative, Training Steps=5M2026.03 | 2.149 | 77.41 | 3.281 | 3.729 | |
| RNDVoC-nonsharedSE Model Type=Discriminative, Training Steps=1M2026.03 | 2.127 | 81.62 | 3.266 | 3.678 | |
| VocosSE Model Type=Discriminative, Training Steps=1M2026.03 | 2.06 | 80.48 | 3.29 | 3.685 | |
| BigVGAN-baseSE Model Type=Discriminative, Training Steps=5M2026.03 | 2.051 | 76.62 | 3.252 | 3.698 | |
| APNet2SE Model Type=Discriminative, Training Steps=1M2026.03 | 1.976 | 76.6 | 3.348 | 3.521 | |
| BigVGANSE Model Type=Generative, Training Steps=5M2026.03 | 1.915 | 73.37 | 3.409 | 3.402 | |
| RNDVoC-sharedSE Model Type=Generative, Training Steps=1M2026.03 | 1.91 | 77.41 | 3.429 | 3.383 | |
| HiFiGAN-V1SE Model Type=Discriminative, Training Steps=1M2026.03 | 1.903 | 75.89 | 3.268 | 3.601 | |
| AvocodoSE Model Type=Discriminative, Training Steps=1M2026.03 | 1.892 | 75.47 | 3.212 | 3.613 | |
| RNDVoC-nonsharedSE Model Type=Generative, Training Steps=1M2026.03 | 1.887 | 77.11 | 3.404 | 3.361 | |
| BigVGAN-baseSE Model Type=Generative, Training Steps=5M2026.03 | 1.839 | 72.77 | 3.422 | 3.375 | |
| VocosSE Model Type=Generative, Training Steps=1M2026.03 | 1.829 | 75.97 | 3.467 | 3.373 | |
| iSTFTNet-V1SE Model Type=Discriminative, Training Steps=1M2026.03 | 1.818 | 74.76 | 3.275 | 3.551 | |
| APNetSE Model Type=Discriminative, Training Steps=1M2026.03 | 1.8 | 77.31 | 3.176 | 3.595 | |
| APNet2SE Model Type=Generative, Training Steps=1M2026.03 | 1.762 | 72.56 | 3.46 | 3.301 | |
| HiFiGAN-V1SE Model Type=Generative, Training Steps=1M2026.03 | 1.739 | 71.78 | 3.44 | 3.318 | |
| AvocodoSE Model Type=Generative, Training Steps=1M2026.03 | 1.72 | 71.48 | 3.389 | 3.311 | |
| iSTFTNet-V1SE Model Type=Generative, Training Steps=1M2026.03 | 1.677 | 70.87 | 3.416 | 3.268 | |
| APNetSE Model Type=Generative, Training Steps=1M2026.03 | 1.63 | 73.19 | 3.335 | 3.29 | |
| MixtureSE Model Type=Discriminative, Training Steps=1M2026.03 | 1.163 | 57.48 | 2.58 | 2.512 | |
| MixtureSE Model Type=Generative, Training Steps=1M2026.03 | 1.163 | 27.48 | 2.58 | 2.512 |