Text-to-Speech on ZTTS1-Eval In-The-Wild (ru)
3.75WERCartesia Sonic 3.5
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Cartesia Sonic 3.52026.06 | 3.75 | 2.64 | 78.2 | |
| Gemini 3.1 Flashsupports zero-shot voice cloning=false2026.06 | 3.77 | 3.36 | 20.9 | |
| Qwen 3 TTS 1.7B2026.06 | 3.88 | 2.74 | 72.8 | |
| Eleven Labs V3supports zero-shot voice cloning=false2026.06 | 4.28 | 3.26 | 19.7 | |
| ZONOS2 8B Quality ModeQuality Mode=true2026.06 | 4.32 | 2.62 | 67.7 | |
| Fish S2 Pro2026.06 | 5.28 | 2.56 | 72.4 | |
| Inworld TTS 22026.06 | 5.32 | 2.66 | 61.8 | |
| VoxCPM 22026.06 | 5.37 | 2.3 | 75.9 | |
| ZONOS2 8BQuality Mode=false2026.06 | 7.8 | 2.3 | 70.9 | |
| Ground truth2026.06 | 12.38 | 2.23 | 79.2 |