Cross-style Speech Synthesis on DeepASMR-DB Cross-style ASMR → Normal
4.04OI-MOS (ZH)Ground Truth
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Ground Truth2026.01 | 4.04 | 4.21 | |
| DeepASMRArchitecture=End-to-End2026.01 | 3.76 | 3.88 | |
| CosyVoice2VC Model=CosyVoiceVC2026.01 | 3.53 | 3.74 | |
| F5TTSVC Model=CosyVoiceVC2026.01 | 3.37 | 3.84 | |
| CosyVoice2VC Model=SeedVC2026.01 | 2.67 | 3.1 | |
| F5TTSVC Model=SeedVC2026.01 | 2.5 | 3.3 |