Text-to-Speech on Seed-TTS zh (test)
0.0084WEROmniVoice
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OmniVoiceParams.=0.8B, Training Data (hours)=581k Multi.2026.04 | 0.0084 | 0.777 | — | — | — | — | 3.11 | — | — | — | — | |
| OmniVoice-EmiliaParams.=0.8B, Training Data (hours)=100k Emilia2026.04 | 0.0089 | 0.765 | — | — | — | — | 3.05 | — | — | — | — | |
| VoxCPMParams.=0.7B, Training Data (hours)=1800k Multi.2026.04 | 0.0099 | 0.772 | — | — | — | — | 2.94 | — | — | — | — | |
| IndexTTS2Params.=1.7B, Training Data (hours)=55k Multi.2026.04 | 0.0105 | 0.764 | — | — | — | — | 3 | — | — | — | — | |
| CosyVoice3Params.=1.1B, Training Data (hours)=1000k Multi.2026.04 | 0.0114 | 0.778 | — | — | — | — | 3.32 | — | — | — | — | |
| Qwen3-TTSParams.=1.1B, Training Data (hours)=5000k Multi.2026.04 | 0.0115 | 0.766 | — | — | — | — | 3.46 | — | — | — | — | |
| Seed-TTS DiT2024.10 | 0.0118 | 0.809 | — | — | — | — | — | — | — | — | — | |
| Ground-truth2026.04 | 0.0125 | 0.755 | — | — | — | — | 2.78 | — | — | — | — | |
| Ground Truth2026.04 | 0.0125 | — | — | — | — | — | — | — | — | — | — | |
| UniSonateParams=1.3B, Data Scale=50k hrs (S) + 20k hrs (M) + 1.5M clips (E), Sampling Rate=44.1kHz2026.04 | 0.0125 | — | — | — | — | — | — | — | — | — | — | |
| Ground Truth#Hz=-, Data (#hours)=-, PT?=false2025.08 | 0.0125 | 0.75 | — | — | — | — | 2.793 | — | — | — | — | |
| Ground Truth2024.10 | 0.0126 | 0.76 | 0 | 3.72 | — | — | — | — | — | — | — | |
| Vocoder Resynthesized2024.10 | 0.0127 | 0.72 | — | — | — | — | — | — | — | — | — | |
| CosyVoice2#Hz=25, Data (#hours)=167K, PT?=true2025.08 | 0.0129 | 0.757 | — | — | — | — | 3.27 | — | — | — | — | |
| InstructAudioParams=1.3B, Data Scale=50k hrs (S) + 20k hrs (M), Sampling Rate=44.1kHz2026.04 | 0.0135 | — | — | — | — | — | — | — | — | — | — | |
| ZipVoiceParams.=0.1B, Training Data (hours)=100k Emilia2026.04 | 0.014 | 0.751 | — | — | — | — | 3.15 | — | — | — | — | |
| ZipVoiceParams=123M, Data Scale=100k hrs (S), Sampling Rate=44.1kHz2026.04 | 0.014 | — | — | — | — | — | — | — | — | — | — | |
| CosyVoice2Params=618M, Data Scale=167k hrs (S), Sampling Rate=44.1kHz2026.04 | 0.0145 | — | — | — | — | — | — | — | — | — | — | |
| FireRedTTS2024.10 | 0.0151 | 0.63 | -0.49 | 3.28 | — | — | — | — | — | — | — | |
| F5-TTSParams.=0.4B, Training Data (hours)=100K Emilia2026.04 | 0.0153 | 0.75 | — | — | — | — | 2.93 | — | — | — | — | |
| F5-TTSParams=336M, Data Scale=100k hrs (S), Sampling Rate=44.1kHz2026.04 | 0.0153 | — | — | — | — | — | — | — | — | — | — | |
| F5-TTSNFE=322024.10 | 0.0156 | 0.76 | 0.21 | 3.83 | — | — | — | — | — | — | — | |
| F5-TTSNFE=162024.10 | 0.0174 | 0.75 | 0.02 | 3.72 | — | — | — | — | — | — | — | |
| E2-TTSParams=333M, Data Scale=100k hrs (S), Sampling Rate=44.1kHz2026.04 | 0.0191 | — | — | — | — | — | — | — | — | — | — | |
| MiMo-Audio-7B-Instruct2025.12 | 0.0196 | — | — | — | — | — | — | — | — | — | — | |
| E2 TTSNFE=322024.10 | 0.0197 | 0.73 | -0.04 | 3.44 | — | — | — | — | — | — | — | |
| Step-Audio2-mini2025.12 | 0.0213 | — | — | — | — | — | — | — | — | — | — | |
| MaskGCT2024.10 | 0.0227 | 0.774 | — | — | — | — | — | — | — | — | — | |
| MaskGCT#Hz=50, Data (#hours)=101K, PT?=false2025.08 | 0.0228 | 0.765 | — | — | — | — | 2.682 | — | — | — | — | |
| MaskGCTParams.=2.2B, Training Data (hours)=100K Emilia2026.04 | 0.024 | 0.773 | — | — | — | — | 2.63 | — | — | — | — | |
| MaskGCTParams=1B, Data Scale=100k hrs (S), Sampling Rate=44.1kHz2026.04 | 0.024 | — | — | — | — | — | — | — | — | — | — | |
| Vevo2#Hz=12.5, Data (#hours)=101K, 7K, PT?=true2025.08 | 0.0294 | 0.754 | — | — | — | — | 2.997 | — | — | — | — | |
| Cosy Voice2024.10 | 0.031 | 0.75 | -0.06 | 3.54 | — | — | — | — | — | — | — | |
| CosyVoice1Params=416M, Data Scale=170k hrs (S), Sampling Rate=44.1kHz2026.04 | 0.0363 | — | — | — | — | — | — | — | — | — | — | |
| F5-TTS#Hz=-, Data (#hours)=101K, PT?=false2025.08 | 0.0387 | 0.71 | — | — | — | — | 2.671 | — | — | — | — | |
| Vevo2-base#Hz=12.5, Data (#hours)=101K, 7K, PT?=false2025.08 | 0.0463 | 0.752 | — | — | — | — | 3.009 | — | — | — | — | |
| Vevo2-base#Hz=12.5, Data (#hours)=101K, PT?=false2025.08 | 0.0711 | 0.752 | — | — | — | — | 3.007 | — | — | — | — | |
| MiMo-Audio-7B-Instruct2025.12 | 0.1414 | — | — | — | — | — | — | — | — | — | — | |
| Step-Audio2-mini2025.12 | 0.1631 | — | — | — | — | — | — | — | — | — | — | |
| Vevo2-base#Hz=12.5, Data (#hours)=7K, PT?=false2025.08 | 0.1699 | 0.726 | — | — | — | — | 2.902 | — | — | — | — | |
| CosyVoice 32026.02 | 0.71 | — | — | — | — | — | — | — | — | — | — | |
| CosyVoice 3zero-shot=true2026.01 | 0.71 | — | — | — | — | — | — | — | — | — | — | |
| CosyVoice 3Setting=Zero-shot, Aspect=Content Consistency2026.04 | 0.71 | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-TTS-12Hz-1.7B-Basezero-shot=true, frame_rate=12Hz, parameters=1.7B2026.01 | 0.77 | — | — | — | — | — | — | — | — | — | — | |
| MiniMax-Speechzero-shot=true2026.01 | 0.83 | — | — | — | — | — | — | — | — | — | — | |
| MiniMax-SpeechSetting=Zero-shot, Aspect=Content Consistency2026.04 | 0.83 | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-TTS-12Hz-0.6B-Basezero-shot=true, frame_rate=12Hz, parameters=0.6B2026.01 | 0.92 | — | — | — | — | — | — | — | — | — | — | |
| KALL-Ezero-shot=true2026.01 | 0.96 | — | — | — | — | — | — | — | — | — | — | |
| Qwen3.5-Omni-PlusSetting=Zero-shot, Aspect=Content Consistency2026.04 | 0.99 | — | — | — | — | — | — | — | — | — | — | |
| Seed-TTS_RL2026.02 | 1 | — | — | — | — | — | — | — | — | — | — | |
| Seed-TTS RLSetting=Zero-shot, Aspect=Content Consistency2026.04 | 1 | — | — | — | — | — | — | — | — | — | — | |
| IndexTTS 2mode=Zero-shot2026.01 | 1.008 | — | — | — | — | 0.865 | — | — | — | — | — | |
| FireRedTTS-2mode=Zero-shot2026.01 | 1.059 | — | — | — | — | 0.772 | — | — | — | — | — | |
| Qwen3-Omni2026.02 | 1.07 | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-Omni-30B-A3BSetting=Zero-shot, Aspect=Content Consistency2026.04 | 1.07 | — | — | — | — | — | — | — | — | — | — | |
| Qwen3-TTS-25Hz-1.7B-Basezero-shot=true, frame_rate=25Hz, parameters=1.7B2026.01 | 1.1 | — | — | — | — | — | — | — | — | — | — | |
| Seed-TTS_ICL2026.02 | 1.11 | — | — | — | — | — | — | — | — | — | — | |
| Seed-TTS ICLSetting=Zero-shot, Aspect=Content Consistency2026.04 | 1.11 | — | — | — | — | — | — | — | — | — | — | |
| Seed-TTSzero-shot=true2026.01 | 1.12 | — | — | — | — | — | — | — | — | — | — | |
| FireRedTTS 2zero-shot=true2026.01 | 1.14 | — | — | — | — | — | — | — | — | — | — | |
| CosyVoice 3mode=Zero-shot2026.01 | 1.16 | — | — | — | — | 0.872 | — | — | — | — | — | |
| Qwen3-TTS-25Hz-0.6B-Basezero-shot=true, frame_rate=25Hz, parameters=0.6B2026.01 | 1.18 | — | — | — | — | — | — | — | — | — | — | |
| Spark TTS2026.02 | 1.2 | — | — | — | — | — | — | — | — | — | — | |
| Spark TTSzero-shot=true2026.01 | 1.2 | — | — | — | — | — | — | — | — | — | — | |
| Spark TTSSetting=Zero-shot, Aspect=Content Consistency2026.04 | 1.2 | — | — | — | — | — | — | — | — | — | — | |
| Ground Truthmode=Zero-shot2026.01 | 1.254 | — | — | — | — | 0.776 | — | — | — | — | — | |
| ERNIE 5.02026.02 | 1.35 | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-Omni2026.02 | 1.42 | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-Omni-7BSetting=Zero-shot, Aspect=Content Consistency2026.04 | 1.42 | — | — | — | — | — | — | — | — | — | — | |
| IndexTTS 2.5mode=Zero-shot2026.01 | 1.426 | — | — | — | — | 0.848 | — | — | — | — | — | |
| CosyVoice 22026.02 | 1.45 | — | — | — | — | — | — | — | — | — | — | |
| CosyVoice2Params=0.5B2026.02 | 1.45 | — | — | — | — | — | — | — | — | — | — | |
| CosyVoice 2Setting=Zero-shot, Aspect=Content Consistency2026.04 | 1.45 | — | — | — | — | — | — | — | — | — | — | |
| F5-TTS2026.02 | 1.56 | — | — | — | — | — | — | — | — | — | — | |
| F5-TTSzero-shot=true2026.01 | 1.56 | — | — | — | — | — | — | — | — | — | — | |
| F5-TTSSetting=Zero-shot, Aspect=Content Consistency2026.04 | 1.56 | — | — | — | — | — | — | — | — | — | — | |
| Llasa-8Bzero-shot=true2026.01 | 1.59 | — | — | — | — | — | — | — | — | — | — | |
| Covo-Audio-BaseParams=7B2026.02 | 1.73 | — | — | — | — | — | — | — | — | — | — | |
| MiMo-Audio-7B-InstructSetting=Zero-shot, Aspect=Content Consistency2026.04 | 1.96 | — | — | — | — | — | — | — | — | — | — | |
| E2 TTS2026.02 | 1.97 | — | — | — | — | — | — | — | — | — | — | |
| E2 TTSzero-shot=true2026.01 | 1.97 | — | — | — | — | — | — | — | — | — | — | |
| E2 TTSSetting=Zero-shot, Aspect=Content Consistency2026.04 | 1.97 | — | — | — | — | — | — | — | — | — | — | |
| GLM-4-Voice-BaseParams=9B2026.02 | 2.1 | — | — | — | — | — | — | — | — | — | — | |
| MaskGCT2026.02 | 2.27 | — | — | — | — | — | — | — | — | — | — | |
| MaskGCTzero-shot=true2026.01 | 2.27 | — | — | — | — | — | — | — | — | — | — | |
| MaskGCTSetting=Zero-shot, Aspect=Content Consistency2026.04 | 2.27 | — | — | — | — | — | — | — | — | — | — | |
| ERNIE 5.0-Base2026.02 | 3.41 | — | — | — | — | — | — | — | — | — | — | |
| CosyVoiceParams=416M, Data (hrs)=170K Multi., Protocol=Zero-shot2026.06 | — | 0.72 | — | — | — | — | — | — | — | 3.63 | — | |
| CosyVoice 2Params=618M, Data (hrs)=167K Multi., Protocol=Zero-shot2026.06 | — | 0.75 | — | — | — | — | — | — | — | 1.45 | — | |
| CosyVoice22026.06 | — | — | — | — | — | — | — | — | 0.748 | — | 1.45 | |
| DSFlowSteps=1, Training Data=Emilia (95K hours)2026.02 | — | 0.84 | — | — | 4.38 | — | — | — | — | — | — | |
| E2-TTSParams=333M, Data (hrs)=100K Emilia, Protocol=Zero-shot2026.06 | — | 0.73 | — | — | — | — | 2.27 | — | — | 1.97 | — | |
| F5-TTSParams=336M, Data (hrs)=100K Emilia, Protocol=Zero-shot2026.06 | — | 0.75 | — | — | — | — | 2.94 | — | — | 1.55 | — | |
| F5-TTS2026.06 | — | — | — | — | — | — | — | — | 0.741 | — | 1.56 | |
| FlashTTSStage=12026.06 | — | — | — | — | — | — | — | — | 0.718 | — | 1.38 | |
| FlashTTSStage=22026.06 | — | — | — | — | — | — | — | — | 0.699 | — | 1.51 | |
| Ground TruthParams=–, Data (hrs)=–, Protocol=Zero-shot2026.06 | — | 0.75 | — | — | — | — | 2.78 | — | — | 1.25 | — | |
| HoliTok-BaseModeling=Unified spoken language modeling2026.05 | — | — | — | — | — | — | — | 4.4 | 0.66 | — | — | |
| HoliTok-UniteModeling=Unified spoken language modeling2026.05 | — | — | — | — | — | — | — | 1.78 | 0.67 | — | — | |
| Llasa-1BParams=1370M, Data (hrs)=250K Multi., Protocol=Zero-shot2026.06 | — | 0.67 | — | — | — | — | — | — | — | 1.89 | — |