End-to-End Automatic Speech Translation of Audiobooks

Alexandre Bérard (1, 2, 3) , Laurent Besacier (2, 3, 1) , Ali Can Kocabiyikoglu (1, 2) , Olivier Pietquin (4, 5)

1 LIG - Laboratoire d'Informatique de Grenoble
2 GETALP - Groupe d’Étude en Traduction Automatique/Traitement Automatisé des Langues et de la Parole
3 IUF - Institut universitaire de France
4 DeepMind Technologies
5 SEQUEL - Sequential Learning

We investigate end-to-end speech-to-text translation on a corpus of audiobooks specifically augmented for this task. Previous works investigated the extreme case where source language transcription is not available during learning nor decoding , but we also study a midway case where source language transcription is available at training time only. In this case, a single model is trained to decode source speech into target text in a single pass. Experimental results show that it is possible to train compact and efficient end-to-end speech translation models in this setup. We also distribute the corpus and hope that our speech translation baseline on this corpus will be challenged in the future.

Mots clés: End-to-end models; Speech Translation; LibriSpeech

Domaines: Informatique et langage [cs.CL]

Alexandre Bérard, Laurent Besacier, Ali Can Kocabiyikoglu, Olivier Pietquin. End-to-End Automatic Speech Translation of Audiobooks. ICASSP 2018 - IEEE International Conference on Acoustics, Speech and Signal Processing, Apr 2018, Calgary, Alberta, Canada. ⟨hal-01709586⟩