Emotion Editing on Emotion Easy Task
1.4WERStep-Audio-EditX-iter0
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Step-Audio-EditX-iter0Parameters=3 B, Training Data (h)=-2026.01 | 1.4 | 0.49 | 3.39 | 50 | |
| Step-Audio-EditX-iter2Parameters=3 B, Training Data (h)=-2026.01 | 1.6 | 0.36 | 3.29 | 57.5 | |
| Step-Audio-EditX-iter1Parameters=3 B, Training Data (h)=-2026.01 | 1.7 | 0.42 | 3.34 | 56.25 | |
| IndexTTS 2Parameters=1.5 B, Training Data (h)=1352026.01 | 2.5 | 0.44 | 3.72 | 56.25 | |
| SpeechEdit-C1Parameters=0.5 B, Training Data (h)=1292026.01 | 2.5 | 0.45 | 4.01 | 63.75 | |
| SpeechEdit-Ablation-Data-C1Parameters=0.5 B, Training Data (h)=1292026.01 | 3.2 | 0.4 | 4 | 60 | |
| SpeechEdit-C2Parameters=0.5 B, Training Data (h)=1292026.01 | 3.9 | 0.37 | 3.98 | 78.75 | |
| CosyVoice 2Parameters=0.5 B, Training Data (h)=<1.5 k2026.01 | 4.1 | 0.52 | 4.01 | 43.75 | |
| SpeechEdit-Ablation-Data-C2Parameters=0.5 B, Training Data (h)=1292026.01 | 5.1 | 0.3 | 3.93 | 76.25 | |
| SpeechEdit-C3Parameters=0.5 B, Training Data (h)=1292026.01 | 6.8 | 0.25 | 4 | 91.25 | |
| SpeechEdit-Ablation-Data-C3Parameters=0.5 B, Training Data (h)=1292026.01 | 9 | 0.21 | 3.89 | 82.5 |