Speech Separation On Libri5Mix
Metrics
SI-SDRi
Results
Performance results of various models on this benchmark
Model Name | SI-SDRi | Paper Title | Repository |
|---|---|---|---|
| Hungarian PIT | 12.72 | Many-Speakers Single Channel Speech Separation with Optimal Permutation Training | |
| Separate And Diffuse | 14.2 | Separate And Diffuse: Using a Pretrained Diffusion Model for Improving Source Separation | - |
| SepIt | 13.7 | SepIt: Approaching a Single Channel Speech Separation Bound | - |
| OCD | 13.4 | OCD: Learning to Overfit with Conditional Diffusion Models |
0 of 4 row(s) selected.