Text-to-Speech on CV3-Eval Cross-lingual en → zh
8.01WERCosyVoice 3
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CosyVoice 3Number of Parameters=1.5B2026.06 | 8.01 | 66.9 | |
| dots.ttsTraining Stage=MF, NFE (Number of Function Evaluations)=42026.06 | 10.73 | 73.8 | |
| dots.ttsTraining Stage=SOAR2026.06 | 10.75 | 75 | |
| dots.ttsTraining Stage=Pretrain2026.06 | 10.88 | 74.6 | |
| CosyVoice 22026.06 | 13.5 | 63.3 |