Audio-Visual Speech Recognition on LRS3 Music noise
7.11CER (-10 dB)VisG AV-HuBERT Large
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| VisG AV-HuBERT LargeBackbone=Large, Noise Type=Music2026.04 | 7.11 | 3.35 | 1.62 | 1.31 | 0.98 | |
| AV-HuBERT LargeBackbone=Large, Noise Type=Music2026.04 | 7.76 | 3.61 | 1.8 | 1.31 | 1.1 | |
| VisG AV-HuBERT BaseBackbone=Base, Noise Type=Music2026.04 | 11.49 | 5.86 | 3.67 | 2.64 | 2.24 | |
| AV-HuBERT BaseBackbone=Base, Noise Type=Music2026.04 | 12.22 | 6.33 | 3.66 | 2.62 | 2.26 |