Speech Separation on WSJ0-5mix
Metric: SI-SDRi (higher is better)
Metric: SI-SDRi (higher is better)
| # | Model↕ | SI-SDRi▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | SepTDA | 21 | No | Boosting Unknown-number Speaker Separation with ... | 2024-01-23 | - |
| 2 | Hungarian PIT | 13.22 | No | Many-Speakers Single Channel Speech Separation w... | 2021-04-18 | Code |
| 3 | Conditional TasNet | 11.7 | No | Sequence to Multi-Sequence Learning via Conditio... | 2020-06-25 | - |
| 4 | TasTas | 11.14 | No | - | - | Code |
| 5 | Gated DualPathRNN | 10.56 | No | Voice Separation with an Unknown Number of Multi... | 2020-02-29 | Code |
| 6 | Multi-Decoder DPRNN | 5.9 | No | Multi-Decoder DPRNN: High Accuracy Source Counti... | 2020-11-24 | Code |