| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Text-to-Speech | CosyVoice zh 3 (test) | UTMOS3.328 | 10 | |
| Text-to-Speech | CosyVoice en 3 (test) | UTMOS3.931 | 10 | |
| Fine-grained Score Accuracy | CosyVoice 2 | Exact Accuracy64.5 | 1 | |
| Binary classification (Human vs Machine speech) | CosyVoice2 Pseudo Human OOD (test) | Accuracy98.44 | 1 |