Text-to-Speech on LibriTTS-R (test-clean)
1.95CERQVD
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| QVDZero-shot=true2025.09 | 1.95 | 5.42 | 4.27 | 0.58 | 0.82 | 0.97 | 0.15 | |
| QVDFine-tuned=true2025.09 | 2.19 | 6.02 | 4.37 | 0.65 | 0.87 | 1.19 | 0.32 | |
| GT2025.09 | 2.41 | 6.17 | 4.17 | 0.81 | — | — | — | |
| VIC-srfBackbone=VITS2025.09 | 2.99 | 7.72 | 4.26 | 0.72 | 0.36 | 0.41 | 0.05 | |
| VIC-disBackbone=VITS2025.09 | 3.12 | 7.99 | 4.25 | 0.75 | 0.37 | 0.51 | 0.14 | |
| VIC-baseBackbone=VITS2025.09 | 3.2 | 8.17 | 4.23 | 0.77 | 0.39 | 0.61 | 0.22 | |
| VITSBackbone=VITS2025.09 | 3.36 | 8.63 | 4.2 | 0.77 | — | — | — |