Speech Editing (Substitution) on Ming-Freeform-Audio-Edit English (full)
3.13DNSMOSCosyEdit
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| CosyEditsettings=recommended2026.05 | 3.13 | 2.56 | 83 | — | 0.146 | |
| ECPAsource=original paper2026.05 | 3.11 | 4.41 | 78 | — | — | |
| SSR-Speech2026.05 | 3.08 | 1.95 | 90 | 0.14 | — | |
| SSR-Speechsettings=recommended2026.05 | 3.08 | 1.95 | 90 | — | 0.14 | |
| GroundTruth2026.05 | 3.05 | — | — | — | — | |
| VoiceCraft-X2026.05 | 3.05 | 6.73 | 81 | 0.146 | — | |
| CosyEdit22026.05 | 3.05 | 1.52 | 90 | 0.132 | — | |
| Ground Truth2026.05 | 3.05 | — | — | — | — | |
| VoiceCraft-Xsettings=recommended2026.05 | 3.05 | 6.73 | 81 | — | 0.146 | |
| LEMAS-Editsettings=recommended2026.05 | 3.05 | 6.56 | 80 | — | 0.161 | |
| CosyEdit2settings=recommended2026.05 | 3.05 | 1.52 | 90 | — | 0.132 | |
| Ming-UniAudio2026.05 | 3.03 | 7.95 | 77 | 0.178 | — | |
| Ming-UniAudiosettings=recommended2026.05 | 3.03 | 7.95 | 77 | — | 0.178 | |
| VoiceCraftsettings=recommended2026.05 | 3.02 | 2.96 | 81 | — | — |