Listen and Translate: A Proof of Concept for End-to-End Speech-to-Text Translation

Abstract : This paper proposes a first attempt to build an end-to-end speech-to-text translation system, which does not use source language text during learning or decoding. Relaxing the need for source language transcription would drastically change the data collection methodology in speech translation, especially in under-resourced scenarios.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [15 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01408086
Contributor : Laurent Besacier <>
Submitted on : Tuesday, December 6, 2016 - 4:50:50 PM
Last modification on : Thursday, April 4, 2019 - 10:18:05 AM
Document(s) archivé(s) le : Tuesday, March 21, 2017 - 1:04:36 PM

File

listen_and_translate_nips_2016...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01408086, version 1

Citation

Alexandre Bérard, Olivier Pietquin, Laurent Besacier, Christophe Servan. Listen and Translate: A Proof of Concept for End-to-End Speech-to-Text Translation. NIPS Workshop on end-to-end learning for speech and audio processing, Dec 2016, Barcelona, Spain. 2016. 〈hal-01408086〉

Share

Metrics

Record views

826

Files downloads

600