Zero-shot Voice Imitation on SeedTTS vc-en (test)
3.31UTMOSMimicLM (SFT)
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| MimicLM (SFT)System Category=Full Voice Imitation (Timbre + Style Transfer), Training Protocol=SFT2026.04 | 3.31 | 4.12 | 4.43 | 4.42 | 12.8 | 0.571 | 0.692 | 0.912 | |
| MimicLM (DPO)System Category=Full Voice Imitation (Timbre + Style Transfer), Training Protocol=DPO2026.04 | 3.22 | 4.15 | 4.45 | 4.45 | 8.25 | 0.601 | 0.699 | 0.925 | |
| CosyVoice 2.0System Category=Timbre-only Voice Conversion2026.04 | 3.04 | 3.98 | 4.31 | 4.38 | 4.28 | 0.539 | 0.647 | 0.919 | |
| SeedVC v2System Category=Full Voice Imitation (Timbre + Style Transfer)2026.04 | 2.94 | 3.65 | 4.14 | 4.01 | 6.32 | 0.553 | 0.653 | 0.917 | |
| MeanVCSystem Category=Timbre-only Voice Conversion2026.04 | 2.84 | 3.65 | 4.04 | 4.14 | 17.82 | 0.419 | 0.45 | 0.909 | |
| LSCodecSystem Category=Timbre-only Voice Conversion2026.04 | 2.84 | 3.94 | 4.33 | 4.25 | 8.76 | 0.445 | 0.583 | 0.911 | |
| VevoSystem Category=Full Voice Imitation (Timbre + Style Transfer)2026.04 | 2.83 | 3.77 | 4.27 | 4 | 9.1 | 0.652 | 0.727 | 0.926 | |
| SeedVCSystem Category=Timbre-only Voice Conversion2026.04 | 2.79 | 3.71 | 4.19 | 4.03 | 3.25 | 0.587 | 0.684 | 0.922 | |
| OpenVoice v2System Category=Timbre-only Voice Conversion2026.04 | 2.55 | 4.15 | 4.47 | 4.43 | 4.15 | 0.424 | 0.552 | 0.918 | |
| FACodecSystem Category=Timbre-only Voice Conversion2026.04 | 2.13 | 3.65 | 4.22 | 3.87 | 4.86 | 0.372 | 0.571 | 0.912 |