Lipreading On Cmlr
Metrics
CER
Results
Performance results of various models on this benchmark
Model Name | CER | Paper Title | Repository |
|---|---|---|---|
| LipCH-Net | 34.07% | A Cascade Sequence-to-Sequence Model for Chinese Mandarin Lip Reading | - |
| CTC/Attention | 9.1% | Visual Speech Recognition for Multiple Languages in the Wild | |
| CSSMCM | 32.48% | A Cascade Sequence-to-Sequence Model for Chinese Mandarin Lip Reading | - |
| WAS | 38.93% | A Cascade Sequence-to-Sequence Model for Chinese Mandarin Lip Reading | - |
| LIBS | 31.27% | Hearing Lips: Improving Lip Reading by Distilling Speech Recognizers |
0 of 5 row(s) selected.