Text-to-Speech on ZTTS1-Eval In-The-Wild (zh)
0.99WERQwen 3 TTS 1.7B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen 3 TTS 1.7B2026.06 | 0.99 | 2.9 | 75.3 | |
| Cartesia Sonic 3.52026.06 | 1.17 | 2.76 | 79.1 | |
| Eleven Labs V3supports zero-shot voice cloning=false2026.06 | 1.22 | 3.36 | 29.7 | |
| Fish S2 Pro2026.06 | 1.26 | 2.73 | 75.5 | |
| Inworld TTS 22026.06 | 1.42 | 2.71 | 66.4 | |
| VoxCPM 22026.06 | 1.44 | 2.4 | 78.7 | |
| Gemini 3.1 Flashsupports zero-shot voice cloning=false2026.06 | 1.45 | 3.55 | 29.9 | |
| ZONOS2 8B Quality ModeQuality Mode=true2026.06 | 2.77 | 2.68 | 70.6 | |
| ZONOS2 8BQuality Mode=false2026.06 | 3.19 | 2.43 | 74.3 | |
| Ground truth2026.06 | 7.78 | 2.39 | 82.1 |