Text-to-speech on Seed-TTS (eval)
1.85WERVoxCPM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VoxCPMBitrate Control=false, Architecture=NAR / Continuous AR2026.02 | 1.85 | 72.9 | |
| CAT-TTSBitrate Control=true, Architecture=Discrete AR2026.02 | 1.89 | 73.1 | |
| Ground Truth2026.01 | 1.9 | — | |
| GLM-TTSBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.02 | 1.91 | 68.1 | |
| OpenAudio-s1-miniBitrate Control=false, Architecture=Discrete AR2026.02 | 1.94 | 55 | |
| FireRedTTS2Bitrate Control=false, Architecture=Discrete AR2026.02 | 1.95 | 66.5 | |
| SparkTTSBitrate Control=false, Architecture=Discrete AR2026.02 | 1.98 | 58.4 | |
| F5-TTSBitrate Control=false, Architecture=NAR / Continuous AR2026.02 | 2 | 67 | |
| CosyVoice 22026.01 | 2.1 | 0.66 | |
| CosyVoice3-1.5BBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.02 | 2.22 | 72 | |
| IndexTTS2Bitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.02 | 2.23 | 70.6 | |
| HiggsAudio-v2Bitrate Control=false, Architecture=Discrete AR2026.02 | 2.44 | 67.7 | |
| MaskGCTBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.02 | 2.62 | 71.7 | |
| FlexiCodec-TTSBitrate Control=true, Architecture=Cascade (AR+NAR / NAR+NAR)2026.02 | 2.63 | 65.7 | |
| Qwen2.5-OmniBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.02 | 2.72 | 63.2 | |
| LlasaBitrate Control=false, Architecture=Discrete AR2026.02 | 2.97 | 57.4 | |
| VibeVoiceBitrate Control=false, Architecture=NAR / Continuous AR2026.02 | 3.04 | 68.9 | |
| CosyVoice2Bitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.02 | 3.09 | 65.9 | |
| Step-Audio2-mini2025.12 | 3.18 | — | |
| FireRedTTSBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.02 | 3.84 | 46 | |
| KanadeToken Rate=25Hz2026.01 | 4 | 0.48 | |
| KMToken Rate=25Hz2026.01 | 4.9 | 0.42 | |
| MiMo-Audio-7B-Instruct2025.12 | 5.37 | — | |
| KMToken Rate=12.5Hz2026.01 | 5.4 | 0.42 | |
| DualCodec2026.01 | 5.5 | 0.34 | |
| KanadeToken Rate=12.5Hz2026.01 | 5.7 | 0.47 | |
| Mimi2026.01 | 6 | 0.32 | |
| SODA-base-4BScale=4B, Shot count=0-shot2026.02 | 6.1 | 0.56 | |
| SODA-base-1.7BScale=1.7B, Shot count=0-shot2026.02 | 6.9 | 0.56 | |
| X-Codec 22026.01 | 7.2 | 0.35 | |
| BiCodec2026.01 | 7.5 | 0.46 | |
| SODA-base-600MScale=600M, Shot count=0-shot2026.02 | 7.6 | 0.555 | |
| PAST2026.01 | 9 | 0.35 | |
| SODA-prelim-600MScale=600M, Shot count=0-shot2026.02 | 9.2 | 0.516 | |
| StableCodec2026.01 | 10.9 | 0.23 | |
| ST2026.01 | 11.2 | 0.35 | |
| SODA-base-135MScale=135M, Shot count=0-shot2026.02 | 11.2 | 0.5 | |
| TiCodec2026.01 | 12.9 | 0.33 | |
| WavTokenizer2026.01 | 15.6 | 0.28 |