Text-To-Speech on English (test)
0.0165WERMiniMax-Speech
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MiniMax-SpeechParams=-, Open source=No2025.12 | 0.0165 | 69.2 | |
| DiTARParams=0.6B, Open source=No2025.12 | 0.0169 | 73.5 | |
| VoxCPMParams=0.5B, Open source=Yes2025.12 | 0.0185 | 72.9 | |
| GLM-TTS_RLParams=1.5B, Open source=Yes2025.12 | 0.0191 | 68.1 | |
| OpenAudio-s1-miniParams=0.5B, Open source=Yes2025.12 | 0.0194 | 55 | |
| FireRedTTS-2Params=-, Open source=Yes2025.12 | 0.0195 | 66.5 | |
| F5-TTSParams=0.3B, Open source=Yes2025.12 | 0.02 | 67 | |
| CosyVoice3Params=0.5B, Open source=Yes2025.12 | 0.0202 | 71.8 | |
| CosyVoice3Params=1.5B, Open source=No2025.12 | 0.0222 | 72 | |
| IndexTTS 2Params=1.5B, Open source=Yes2025.12 | 0.0223 | 70.6 | |
| GLM-TTSParams=1.5B, Open source=Yes2025.12 | 0.0223 | 67.2 | |
| Seed-TTSParams=-, Open source=No2025.12 | 0.0225 | 76.2 | |
| HiggsAudio-v2Params=3B, Open source=Yes2025.12 | 0.0244 | 67.7 | |
| MaskGCTParams=-, Open source=Yes2025.12 | 0.0262 | 71.7 | |
| Qwen2.5-OmniParams=7B, Open source=Yes2025.12 | 0.0272 | 63.2 | |
| MegaTTS3Params=0.5B, Open source=No2025.12 | 0.0279 | 77.1 | |
| VibeVoiceParams=1.5B, Open source=Yes2025.12 | 0.0304 | 68.9 | |
| CosyVoice2Params=0.5B, Open source=Yes2025.12 | 0.0309 | 65.9 | |
| SparkTTSParams=0.5B, Open source=Yes2025.12 | 0.0314 | 57.3 | |
| FireRedTTSParams=0.5B, Open source=Yes2025.12 | 0.0382 | 46 | |
| CosyVoiceParams=0.3B, Open source=Yes2025.12 | 0.0429 | 60.9 |