Controllable Speech Synthesis on Prosody-timbre Evaluation Set Style scenario
51.5DisCo-Speech PreferenceDisCo-Speech
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| DisCo-SpeechAspect=Timbre, Comparison Baseline=Vevo2025.12 | 51.5 | 27.2 | 21.3 | |
| DisCo-SpeechAspect=Prosody, Comparison Baseline=Vevo2025.12 | 48.9 | 31.1 | 20 | |
| DisCo-SpeechAspect=Prosody, Comparison Baseline=IndexTTS 22025.12 | 41.7 | 25.3 | 33 | |
| DisCo-SpeechAspect=Timbre, Comparison Baseline=IndexTTS 22025.12 | 37.4 | 33.6 | 29 |