Emotional Text-to-Speech on ESD
75Score (Angry)DUET + GradTTS
Evaluation results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DUET + GradTTSPlug-and-play=true, Frozen backbones=true2026.05 | 75 | — | — | — | — | — | — | — | — | 73.8 | 77.8 | 75.5 | |
| DUET + ProDiffPlug-and-play=true, Frozen backbones=true2026.05 | 53.5 | — | — | — | — | — | — | — | — | 89.7 | 47.5 | 63.6 | |
| EmoVoice2026.05 | 50.7 | — | — | — | — | — | — | — | — | 13 | 27 | 30.2 | |
| DUET + F5-TTSPlug-and-play=true, Frozen backbones=true, Reference speech with neutral emotion at inference=true2026.05 | 40.9 | — | — | — | — | — | — | — | — | 75.2 | 78.7 | 64.9 | |
| OpenAudio2026.05 | 33.5 | — | — | — | — | — | — | — | — | 38.8 | 2.8 | 25 | |
| DUET + StableTTSPlug-and-play=true, Frozen backbones=true2026.05 | 29.5 | — | — | — | — | — | — | — | — | 47.8 | 52.5 | 43.3 | |
| Qwen3-TTS2026.05 | 27.7 | — | — | — | — | — | — | — | — | 71.5 | 41.3 | 46.8 | |
| DUET + MatchaPlug-and-play=true, Frozen backbones=true2026.05 | 26.3 | — | — | — | — | — | — | — | — | 88.5 | 78.2 | 64.3 | |
| EmoKnobReference speech with neutral emotion at inference=true2026.05 | 25.2 | — | — | — | — | — | — | — | — | 81.3 | 16.7 | 41.1 | |
| CosyVoice2Reference speech with neutral emotion at inference=true2026.05 | 21.3 | — | — | — | — | — | — | — | — | 43.2 | 59.3 | 41.3 | |
| ChatTTS2026.05 | 13.2 | — | — | — | — | — | — | — | — | 39.1 | 32.2 | 28.2 | |
| IndexTTS22026.05 | 12.7 | — | — | — | — | — | — | — | — | 42.2 | 25.7 | 26.9 | |
| Chatterbox2026.05 | 6.2 | — | — | — | — | — | — | — | — | 18 | 10.8 | 11.7 | |
| EmotiVoice2026.05 | 5 | — | — | — | — | — | — | — | — | 30 | 3.3 | 12.8 | |
| EmoSphere++2026.05 | 0 | — | — | — | — | — | — | — | — | 28.6 | 33.9 | 20.8 | |
| ExpProModel=CosyVoice2024.09 | — | 3.87 | 0.689 | 3.8 | 0.651 | 3.86 | 0.633 | 3.93 | 0.681 | — | — | — | |
| ExpProModel=GPT-SoVITS2024.09 | — | 3.78 | 0.659 | 3.75 | 0.641 | 3.65 | 0.611 | 3.88 | 0.675 | — | — | — | |
| MiniLMModel=CosyVoice2024.09 | — | 3.79 | 0.681 | 3.75 | 0.647 | 3.81 | 0.628 | 3.88 | 0.675 | — | — | — | |
| MiniLMModel=GPT-SoVITS2024.09 | — | 3.71 | 0.637 | 3.62 | 0.628 | 3.51 | 0.591 | 3.77 | 0.671 | — | — | — | |
| RandomModel=CosyVoice2024.09 | — | 3.83 | 0.683 | 3.71 | 0.635 | 3.84 | 0.621 | 3.83 | 0.667 | — | — | — | |
| RandomModel=GPT-SoVITS2024.09 | — | 3.69 | 0.641 | 3.57 | 0.597 | 3.59 | 0.583 | 3.72 | 0.641 | — | — | — |