Zero-shot Text-to-Speech on CV3-Eval Multi-lingual Voice Cloning Chinese
4.64IMOSCosyEdit2
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CosyEdit2Setting=Zero-shot2026.05 | 4.64 | 4.512 | |
| CosyVoice2Setting=Zero-shot2026.05 | 4.535 | 4.432 | |
| CosyEditSetting=Zero-shot2026.05 | 4.275 | 4.272 | |
| VoiceCraft-XSetting=Zero-shot2026.05 | 3.882 | 3.53 |