Visual Speech Recognition on Multilingual TEDx Italian (MTit) (test)
57.9Mean WER
VSR model with prediction-based auxiliary tasks
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| 57.9 | 0.7 | 57.4 | ||
| 58.7 | 0.3 | 58.2 | ||
| 65.9 | 0.5 | 65.2 | ||
2022.02 | 71.5 | 0.4 | 70.9 |
VSR model with prediction-based auxiliary tasks
| Method | Links | |||
|---|---|---|---|---|
| 57.9 | 0.7 | 57.4 | ||
| 58.7 | 0.3 | 58.2 | ||
| 65.9 | 0.5 | 65.2 | ||
2022.02 | 71.5 | 0.4 | 70.9 |