Video-to-Speech Synthesis on V2C Dub 3.0
3.94MOS-SStyleDubber
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| StyleDubberVisual=true2024.02 | 3.94 | 3.87 | 71.52 | 13.45 | |
| Zero-shot TTSVisual=true2024.02 | 3.62 | 3.31 | 61.12 | 19.25 | |
| Zero-shot TTSVisual=false2024.02 | 3.53 | 3.35 | 57.23 | 19.83 | |
| StyleSpeechVisual=true2024.02 | 3.31 | 3.22 | 58.71 | 89.11 | |
| StyleSpeechVisual=false2024.02 | 3.17 | 3.22 | 55.81 | 77.46 | |
| HPMDubbingVisual=true2024.02 | 3.11 | 2.92 | 49.31 | 106.81 | |
| Face-TTSVisual=true2024.02 | 3.1 | 3.17 | 33.8 | 201.98 | |
| V2C-NetVisual=true2024.02 | 3.05 | 2.83 | 38.43 | 112.71 | |
| Fastspeech2Visual=true2024.02 | 2.97 | 3.03 | 26.79 | 18.38 | |
| Fastspeech2Visual=false2024.02 | 2.91 | 3.02 | 21.11 | 14.05 |