Zero-shot Voice Cloning on Seed-TTS-Eval Average
2.92WER (%)dots.tts (Pretrain)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| dots.tts (Pretrain)Params=2B, Variant=Pretrain2026.06 | 2.92 | 78.8 | |
| dots.tts (MF, NFE=4)Params=2B, Distillation Method=MeanFlow, NFE=42026.06 | 2.94 | 78.2 | |
| dots.tts (SOAR)Params=2B, Variant=SOAR2026.06 | 2.95 | 79.2 | |
| CosyVoice 3Params=1.5B2026.06 | 3.06 | 75.3 | |
| Qwen3-TTSParams=1.7B2026.06 | 3.07 | 74.5 | |
| dots.tts (MF, NFE=3)Params=2B, Distillation Method=MeanFlow, NFE=32026.06 | 3.21 | 78.1 | |
| dots.tts (MF, NFE=2)Params=2B, Distillation Method=MeanFlow, NFE=22026.06 | 3.43 | 77 | |
| IndexTTS 2Params=1.5B2026.06 | 3.46 | 74.2 | |
| Seed-TTS2026.06 | 3.65 | 77.8 | |
| VoxCPM 2Params=2B2026.06 | 3.65 | 76.7 | |
| FireRedTTS 2Params=1.5B2026.06 | 4.02 | 70.1 | |
| F5-TTSParams=0.3B2026.06 | 4.1 | 71.4 |