Speech To Text Translation On Libri Trans
Métriques
Case-insensitive sacreBLEU
Case-insensitive tokenized BLEU
Case-sensitive sacreBLEU
Case-sensitive tokenized BLEU
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | Case-insensitive sacreBLEU | Case-insensitive tokenized BLEU | Case-sensitive sacreBLEU | Case-sensitive tokenized BLEU | Paper Title | Repository |
|---|---|---|---|---|---|---|
| Transformer + ASR Pretrain + SpecAug | 17.2 | 18.7 | 16.3 | 17.8 | NeurST: Neural Speech Translation Toolkit | |
| Transformer + ASR Pretrain | 16.5 | 17.9 | 15.5 | 16.9 | NeurST: Neural Speech Translation Toolkit |
0 of 2 row(s) selected.