Speech Recognition on Turkish video dataset (clean)
16.1WER+ crosslingual pretrain + aux + kl loss
Evaluation results
| Method | Links | ||
|---|---|---|---|
| + crosslingual pretrain + aux + kl losslambda_aux=0.32020.11 | 16.1 | — | |
| + ce pretrain + ce loss, mid, toplambda_ce=0.62020.11 | 16.3 | — | |
| + aux + kl losslambda_aux=0.32020.11 | 16.4 | — | |
| + ce pretrain + ce loss, midlambda_ce=0.62020.11 | 16.5 | — | |
| + crosslingual pretrain2020.11 | 16.6 | — | |
| + kl losslambda_aux=0.32020.11 | 16.7 | — | |
| + aux losslambda_aux=0.32020.11 | 16.8 | — | |
| + ce pretrain2020.11 | 16.8 | — | |
| baseline2020.11 | 17.1 | — |