Text-to-Speech on Balalaika corpus (test)
2.938MOS (Mean Opinion Score)F5
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| F5model=F52026.03 | 2.938 | 3.158 | 3.0166 | 4.2379 | 4.2286 | 4.4381 | 4.1848 | 4.2128 | 17.91 | 5.56 | 93.5 | |
| XTTSmodel=XTTS2026.03 | 2.905 | 3.195 | 3.0509 | 3.6738 | 3.8541 | 4.1179 | 3.8621 | 3.6105 | 8.35 | 3.71 | 86.1 | |
| DiTTo-TTSconfiguration=1+2 encoder2026.03 | 2.521 | 3.098 | 2.8078 | 4.3426 | 4.2362 | 4.2506 | 4.2275 | 4.3895 | 12.54 | 7.04 | 83.8 | |
| Tortoisemodel=Tortoise2026.03 | 2.51 | 2.812 | 2.9567 | 3.285 | 4.0453 | 3.7156 | 3.5483 | 4.0243 | 10.83 | 3.57 | 73.49 |