Voice Conversion on LibriTTS (test-clean)
2.04WERGT
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GTEvaluation Protocol=Ground Truth2024.10 | 2.04 | 4.17 | — | 4.21 | — | — | |
| Takin-VCEvaluation Protocol=zero-shot, Steps=10, CFG=0.7, RTF=0.1542024.10 | 2.35 | 3.98 | 4.11 | 4.08 | 0.71 | — | |
| SeedVCEvaluation Protocol=zero-shot, Steps=10, CFG=0.7, RTF=0.3412024.10 | 2.51 | 3.87 | 3.74 | 3.81 | 0.68 | — | |
| VALLE-VCEvaluation Protocol=zero-shot, RTF=3.6782024.10 | 2.77 | 3.8 | 3.79 | 3.72 | 0.65 | — | |
| StableVCEvaluation Protocol=zero-shot, Steps=10, CFG=0.7, RTF=0.2672024.10 | 2.77 | 3.83 | 3.88 | 3.92 | 0.66 | — | |
| NS2VCEvaluation Protocol=zero-shot, Steps=10, CFG=0.7, RTF=0.3472024.10 | 2.94 | 3.65 | 3.51 | 3.64 | 0.53 | — | |
| DiffVCEvaluation Protocol=zero-shot, Steps=10, CFG=0.7, RTF=0.2942024.10 | 3.08 | 3.75 | 3.66 | 3.68 | 0.61 | — | |
| SEFVCEvaluation Protocol=zero-shot, RTF=0.1872024.10 | 3.75 | 3.68 | 3.76 | 3.51 | 0.63 | — | |
| DSA-Tokenizer2026.01 | 23.95 | — | — | 3.9 | — | 0.41 | |
| SAC2026.01 | 24.21 | — | — | 3.3 | — | 0.37 | |
| WavTokenizer2026.01 | 89.44 | — | — | 3.75 | — | 0.28 |