Audio-visual Speech Recognition on LRS3 Occlusion by hands
26.6WER (Babble, -10 dB)
CAV2vec
Updated 3d ago
Evaluation Results
| Method | Links | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2025.12 | 26.6 | 12.4 | 4.5 | 2.6 | 1.8 | 9.6 | 6.2 | 3.6 | 2.6 | 2.2 | 1.7 | 3.3 | 9.4 | 4.8 | 2.6 | 1.9 | 1.7 | 4.1 | 5.2 | 7.4 | 1.5 | |
2025.12 | 31.7 | 14.5 | 5.1 | 2.7 | 2 | 11.2 | 7.7 | 4.2 | 2.6 | 2.2 | 1.9 | 3.7 | 11.2 | 5.3 | 3.1 | 2.1 | 1.7 | 4.7 | 6.1 | 8.7 | 1.6 | |
2025.12 | 32 | 15.5 | 5.3 | 2.7 | 2 | 11.5 | 8.1 | 4.2 | 3 | 2.3 | 2 | 3.9 | 11.5 | 5.3 | 2.9 | 2.3 | 1.8 | 4.8 | 6.2 | 9 | 1.6 | |
2025.12 | 33.2 | 15.6 | 5.7 | 2.6 | 2 | 11.8 | 8.1 | 4.7 | 2.8 | 2.3 | 1.9 | 4 | 12.3 | 6 | 2.9 | 2.2 | 1.9 | 5 | 6.5 | 9.4 | 1.5 | |
2025.12 | 38.3 | 21.6 | 5.7 | 2.5 | 2 | 14 | 41.8 | 29 | 13.9 | 4.6 | 2.4 | 18.4 | 18.5 | 7.5 | 3.3 | 2.2 | 1.9 | 6.7 | 11.4 | 17.4 | 1.7 | |
2025.12 | 46 | 25.8 | 9 | 4.2 | 2.8 | 17.6 | 60.6 | 44 | 18.9 | 5.5 | 2.9 | 26.4 | 23.7 | 11.4 | 5.1 | 3.4 | 2.7 | 9.3 | 15.6 | 23.7 | 2.2 | |
2025.12 | 57.2 | 31.9 | 13.7 | 7.5 | 5.3 | 23.1 | 45.7 | 29.3 | 17.1 | 10 | 6.8 | 21.8 | 32.3 | 18 | 9.2 | 6.2 | 5.4 | 14.2 | 18.3 | 26.2 | 4.1 |