Movie Dubbing on CinePile-Dub Out-of-Domain Movie
60.58SPKSIMGT
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GTProtocol=Ground Truth2026.04 | 60.58 | 2.56 | 100 | 0 | 3.15 | 100 | |
| CoSync-DiTProtocol=Zero-shot, Setting=2, Training Dataset=CelebV-Dub, Reference audio source=different segments of the same speaker2026.04 | 47.24 | 7.53 | 70.12 | 0.319 | 3.39 | 31.79 | |
| AlignDiTProtocol=Zero-shot, Setting=2, Training Dataset=CelebV-Dub, Reference audio source=different segments of the same speaker2026.04 | 38.58 | 34.52 | 66.74 | 0.358 | 3.37 | 22.5 | |
| HPMDubbingProtocol=Zero-shot, Setting=2, Training Dataset=CelebV-Dub, Reference audio source=different segments of the same speaker2026.04 | 30.75 | 93.86 | 52.93 | 0.382 | 3.24 | 13.87 | |
| StyleDubberProtocol=Zero-shot, Setting=2, Training Dataset=CelebV-Dub, Reference audio source=different segments of the same speaker2026.04 | 28.67 | 77.61 | 57.15 | 0.372 | 2.88 | 16.48 | |
| ProDubberProtocol=Zero-shot, Setting=2, Training Dataset=CelebV-Dub, Reference audio source=different segments of the same speaker2026.04 | 23.91 | 8.02 | 62.06 | 0.338 | 2.93 | 20.8 | |
| InstructDubProtocol=Zero-shot, Setting=2, Training Dataset=CelebV-Dub, Reference audio source=different segments of the same speaker2026.04 | 23.11 | 7.62 | 61.14 | 0.367 | 2.94 | 16.66 | |
| FlowDubberProtocol=Zero-shot, Setting=2, Training Dataset=CelebV-Dub, Reference audio source=different segments of the same speaker2026.04 | 19.86 | 55.93 | 63.86 | 0.346 | 3.05 | 13.05 | |
| EmoDubberProtocol=Zero-shot, Setting=2, Training Dataset=CelebV-Dub, Reference audio source=different segments of the same speaker2026.04 | 19.43 | 57.73 | 62.46 | 0.347 | 3.05 | 13 | |
| HD-DubberProtocol=Zero-shot, Setting=2, Training Dataset=CelebV-Dub, Reference audio source=different segments of the same speaker2026.04 | 7.04 | 64.12 | 47.66 | 0.33 | 2.58 | 18.86 |