Speech Enhancement on EARS-WHAM
Metric: POLQA (higher is better)
Metric: POLQA (higher is better)
| # | Model↕ | POLQA▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | Schrödinger Bridge (PESQ loss) | 3.71 | No | Investigating Training Objectives for Generative... | 2024-09-16 | Code |
| 2 | Schrödinger Bridge | 3.46 | No | Schrödinger Bridge for Generative Speech Enhance... | 2024-07-22 | - |
| 3 | SGMSE+ | 3.4 | No | Speech Enhancement and Dereverberation with Diff... | 2022-08-11 | Code |
| 4 | Demucs v4 | 2.97 | No | Hybrid Transformers for Music Source Separation | 2022-11-15 | Code |
| 5 | Conv-TasNet | 2.73 | No | Conv-TasNet: Surpassing Ideal Time-Frequency Mag... | 2018-09-20 | Code |
| 6 | CDiffuSE | 1.81 | No | Conditional Diffusion Probabilistic Model for Sp... | 2022-02-10 | Code |