Text-to-Speech on VccmDataset (test A)
0.954PitchGT Codec
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| GT Codec2024.06 | 0.954 | 0.885 | 0.977 | — | 2.6 | 0.96 | — | |
| PromptTTS 2Clone Timbre=false, Control Style=true2024.06 | 0.867 | 0.785 | 0.825 | 0.406 | 3.1 | — | — | |
| InstructTTSClone Timbre=false, Control Style=true2024.06 | 0.849 | 0.761 | 0.822 | 0.412 | 3 | 0.86 | — | |
| ControlSpeechClone Timbre=true, Control Style=true2024.06 | 0.833 | 0.829 | 0.894 | 0.557 | 2.9 | 0.89 | — | |
| PromptStyleClone Timbre=false, Control Style=true2024.06 | 0.831 | 0.786 | 0.787 | 0.366 | 3.3 | 0.84 | — | |
| SalleClone Timbre=false, Control Style=true2024.06 | 0.788 | 0.756 | 0.831 | 0.389 | 5.5 | — | — |