Visual Speech Recognition on LRS3 v0.0 (test)
1.2WEROurs (raw A + V)
Evaluation Results
| Method | Links | |
|---|---|---|
| Ours (raw A + V)Training Data (Hours)=LRW (157) + LRS3v0.0 (474)2021.02 | 1.2 | |
| EG-seq2seqTraining Data (Hours)=LRW (157) + LRS3v0.0 (474)2021.02 | 6.8 | |
| Ours (V)Training Data (Hours)=LRW (157) + LRS3v0.0 (474)2021.02 | 30.4 | |
| EG-seq2seqTraining Data (Hours)=LRW (157) + LRS3v0.0 (474)2021.02 | 57.8 | |
| Conv-seq2seqTraining Data (Hours)=LRW (157) + LRS2&3v0.0 (698)2021.02 | 60.1 |