Video-to-Speech on LRS2 (test)
8.93WER (Word Error Rate)Ground Truth
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Ground TruthVenue=-, Audio guidance (A)=-, Video guidance (V)=-, Trained on target dataset (LRS2)=true2026.04 | 8.93 | 3.14 | 3.05 | 0 | 7.2 | 6.67 | 100 | 1 | |
| FTVVenue=CVPR’25, Audio guidance (A)=✗, Video guidance (V)=✓, Trained on target dataset (LRS2)=true2026.04 | 38.09 | 3.11 | 3.88 | 12.91 | 7.71 | 6.35 | 67.84 | 0.5368 | |
| HiCoDiTVenue=-, Audio guidance (A)=✗, Video guidance (V)=✓, Trained on target dataset (LRS2)=false2026.04 | 39.99 | 3.35 | 3.68 | 8.74 | 7.95 | 6.17 | 68.21 | 0.5222 | |
| HiCoDiTVenue=-, Audio guidance (A)=✓, Video guidance (V)=✓, Trained on target dataset (LRS2)=false2026.04 | 40.75 | 3.27 | 3.38 | 8.36 | 7.83 | 6.24 | 65.65 | 0.5954 | |
| AlignDiTVenue=ACM MM’25, Audio guidance (A)=✗, Video guidance (V)=✓, Trained on target dataset (LRS2)=false2026.04 | 42.26 | 3.13 | 3.65 | 8.46 | 7.5 | 6.58 | 67.01 | 0.5187 | |
| EmoDubberVenue=CVPR’25, Audio guidance (A)=✓, Video guidance (V)=✓, Trained on target dataset (LRS2)=true2026.04 | 47.6 | 2.84 | 2.77 | 7.02 | 7.42 | 6.6 | 66.76 | 0.5252 | |
| DiffV2SVenue=ICCV’23, Audio guidance (A)=✗, Video guidance (V)=✓, Trained on target dataset (LRS2)=true2026.04 | 54.86 | 2.36 | 2.95 | — | — | — | — | — | |
| MTLVenue=ICASSP’23, Audio guidance (A)=✓, Video guidance (V)=✓, Trained on target dataset (LRS2)=true2026.04 | 58.03 | 2.42 | 1.3 | 10.71 | 6.58 | 7.16 | 63.89 | 0.3556 | |
| LTBSVenue=AAAI’24, Audio guidance (A)=✗, Video guidance (V)=✓, Trained on target dataset (LRS2)=false2026.04 | 94.3 | 2.17 | 2.29 | — | — | — | — | — | |
| Lip2WavVenue=CVPR’20, Audio guidance (A)=✓, Video guidance (V)=✓, Trained on target dataset (LRS2)=false2026.04 | 100.05 | 2.47 | 1.31 | 14.09 | 3.83 | 9.8 | 54.38 | 0.4438 |