Controllable Speech Synthesis on Prosody-timbre Evaluation Set Emotion scenario
50.6DisCo-Speech PreferenceDisCo-Speech
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| DisCo-SpeechAspect=Prosody, Comparison Baseline=Vevo2025.12 | 50.6 | 12.7 | 36.7 | |
| DisCo-SpeechAspect=Timbre, Comparison Baseline=Vevo2025.12 | 45.3 | 14.5 | 40.2 | |
| DisCo-SpeechAspect=Timbre, Comparison Baseline=IndexTTS 22025.12 | 42.5 | 23.8 | 33.7 | |
| DisCo-SpeechAspect=Prosody, Comparison Baseline=IndexTTS 22025.12 | 36 | 25.5 | 38.5 |