MVSep Piano model is based on MDX23C, MelRoformer and SCNet Large architectures. It produces high quality separation for piano and other stems. We provide comparison with other public model (Demucs4HT (6 stems)). Used metrics is SDR - the more the better.
See the results in table below.
| Algorithm name | Validation type | |||
| piano (SDR) | other (SDR) | |||
| Demucs4HT (6 stems) | 2.23 | 14.51 | ||
| mdx23c (2023.08, SDR: 4.79) | 4.79 | 17.07 | ||
| mdx23c (2024.09, SDR: 5.59) | 5.59 | 17.89 | ||
| MelRoformer (viperx, SDR: 5.67) | 5.67 | 17.95 | ||
| SCNet Large (2024.09, SDR: 5.89) | 5.89 | 18.16 | ||
| Ensemble (SCNet + Mel, SDR: 6.19) | 6.19 | 18.47 | ||
| BS Roformer SW (SDR: 7.83) | 7.83 | 19.97 | ||
