Text-to-Speech on ZTTS1-Eval In-The-Wild (train)
4.84WERCartesia Sonic 3.5
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Cartesia Sonic 3.52026.06 | 4.84 | 2.57 | 78.5 | |
| VoxCPM 22026.06 | 6.26 | 2.32 | 76.9 | |
| Gemini 3.1 Flashsupports zero-shot voice cloning=false2026.06 | 6.92 | 3.66 | 26.7 | |
| ZONOS2 8B Quality ModeQuality Mode=true2026.06 | 7.47 | 2.64 | 67.7 | |
| Fish S2 Pro2026.06 | 7.64 | 2.58 | 74.2 | |
| Eleven Labs V3supports zero-shot voice cloning=false2026.06 | 7.78 | 3.44 | 26.4 | |
| ZONOS2 8BQuality Mode=false2026.06 | 10.46 | 2.41 | 72.1 | |
| Ground truth2026.06 | 20.86 | 2.27 | 80.8 | |
| Qwen 3 TTS 1.7B2026.06 | 83.2 | 2.7 | 62.1 |