Emotional Text-to-Speech on ESD MED-TTS (Chinese)
0.0311CERSparkTTS
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| SparkTTSPrompt Setting=Speech Emotion Prompt, Language=Chinese2026.01 | 0.0311 | — | 3.466 | 0.382 | 4.338 | 3.345 | 0.807 | — | — | — | |
| CosyVoice2Prompt Setting=Speech Emotion Prompt, Language=Chinese2026.01 | 0.0338 | — | 3.306 | 0.423 | 4.147 | 3.313 | 0.766 | — | — | — | |
| Segment-Aware ConditioningPrompt Setting=Speech Emotion Prompt, Language=Chinese2026.01 | 0.0379 | — | 3.752 | 0.47 | 4.509 | 3.37 | 0.724 | — | — | — | |
| IndexTTS2Prompt Setting=Speech Emotion Prompt, Language=Chinese2026.01 | 0.0402 | — | 3.694 | 0.401 | 4.146 | 3.289 | 0.869 | — | — | — | |
| MaskGCTPrompt Setting=Speech Emotion Prompt, Language=Chinese2026.01 | 0.0722 | — | 3.693 | 0.35 | 4.309 | 3.278 | 0.814 | — | — | — | |
| F5TTSPrompt Setting=Speech Emotion Prompt, Language=Chinese2026.01 | 0.1032 | — | 3.314 | 0.324 | 3.718 | 3.228 | 0.734 | — | — | — | |
| CosyVoice2Prompt Setting=Text Emotion Prompt, Language=Chinese2026.01 | 4.488 | — | 3.105 | 0.477 | 4.346 | 3.206 | 0.222 | — | — | — | |
| Segment-Aware ConditioningPrompt Setting=Text Emotion Prompt, Language=Chinese2026.01 | 5.893 | — | 3.357 | 0.421 | 4.407 | 3.295 | 0.531 | — | — | — | |
| IndexTTS2Prompt Setting=Text Emotion Prompt, Language=Chinese2026.01 | 6.962 | — | 3.212 | 0.369 | 4.179 | 3.169 | 0.702 | — | — | — |