Emotional Text-to-Speech on ESD (English)
4.35SMOSCSP-FT
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| CSP-FTBackbone Model=Fun-CosyVoice3.02025.01 | 4.35 | — | — | — | — | — | — | — | 4.4 | 4.42 | |
| Full FTBackbone Model=Fun-CosyVoice3.02025.01 | 4.28 | — | — | — | — | — | — | — | 3.45 | 4.45 | |
| CSP-FTBackbone Model=CosyVoice2025.01 | 4.18 | — | — | — | — | — | — | — | 4.15 | 4.28 | |
| Full FTBackbone Model=CosyVoice2025.01 | 4.15 | — | — | — | — | — | — | — | 2.85 | 4.3 | |
| LoRA FTBackbone Model=Fun-CosyVoice3.02025.01 | 4.15 | — | — | — | — | — | — | — | 3.95 | 4.25 | |
| LoRA FTBackbone Model=CosyVoice2025.01 | 4.08 | — | — | — | — | — | — | — | 3.12 | 4.15 | |
| CSP-FTBackbone Model=VALLE-X2025.01 | 3.95 | — | — | — | — | — | — | — | 4.2 | 4.12 | |
| OriginBackbone Model=Fun-CosyVoice3.02025.01 | 3.95 | — | — | — | — | — | — | — | 4.35 | 3.88 | |
| Full FTBackbone Model=VALLE-X2025.01 | 3.92 | — | — | — | — | — | — | — | 3.85 | 4.05 | |
| LoRA FTBackbone Model=VALLE-X2025.01 | 3.85 | — | — | — | — | — | — | — | 3.8 | 3.95 | |
| OriginBackbone Model=CosyVoice2025.01 | 3.8 | — | — | — | — | — | — | — | 4.25 | 3.75 | |
| OriginBackbone Model=VALLE-X2025.01 | 3.65 | — | — | — | — | — | — | — | 4.12 | 3.52 | |
| CSP-FTBackbone Model=GPT-SoVITS2025.01 | 3.61 | — | — | — | — | — | — | — | 4.02 | 3.78 | |
| Full FTBackbone Model=GPT-SoVITS2025.01 | 3.55 | — | — | — | — | — | — | — | 3.52 | 3.82 | |
| LoRA FTBackbone Model=GPT-SoVITS2025.01 | 3.48 | — | — | — | — | — | — | — | 3.25 | 3.75 | |
| OriginBackbone Model=GPT-SoVITS2025.01 | 3.12 | — | — | — | — | — | — | — | 4.15 | 3.35 | |
| CosyVoice2Prompt Setting=Speech Emotion Prompt, Language=English2026.01 | — | 1.411 | — | 3.605 | 0.402 | 4.535 | 3.316 | 0.831 | — | — | |
| CosyVoice2Prompt Setting=Text Emotion Prompt, Language=English2026.01 | — | 1.522 | — | 3.465 | 0.453 | 4.33 | 3.271 | 0.303 | — | — | |
| F5TTSPrompt Setting=Speech Emotion Prompt, Language=English2026.01 | — | 2.632 | — | 3.674 | 0.353 | 4.427 | 3.33 | 0.832 | — | — | |
| IndexTTS2Prompt Setting=Speech Emotion Prompt, Language=English2026.01 | — | 2.454 | — | 3.871 | 0.457 | 4.465 | 3.304 | 0.861 | — | — | |
| IndexTTS2Prompt Setting=Text Emotion Prompt, Language=English2026.01 | — | 2.246 | — | 3.543 | 0.424 | 4.299 | 3.216 | 0.525 | — | — | |
| MaskGCTPrompt Setting=Speech Emotion Prompt, Language=English2026.01 | — | 3.52 | — | 3.829 | 0.347 | 4.475 | 3.275 | 0.854 | — | — | |
| Segment-Aware ConditioningPrompt Setting=Speech Emotion Prompt, Language=English2026.01 | — | 2.519 | — | 3.925 | 0.485 | 4.706 | 3.395 | 0.837 | — | — | |
| Segment-Aware ConditioningPrompt Setting=Text Emotion Prompt, Language=English2026.01 | — | 3.038 | — | 3.694 | 0.462 | 4.569 | 3.335 | 0.433 | — | — | |
| SparkTTSPrompt Setting=Speech Emotion Prompt, Language=English2026.01 | — | 2.433 | — | 3.456 | 0.358 | 4.494 | 3.404 | 0.849 | — | — |