Publication Details

The IWSLT 2021 BUT Speech Translation Systems

VYDANA Hari K., KARAFIÁT Martin, BURGET Lukáš and ČERNOCKÝ Jan. The IWSLT 2021 BUT Speech Translation Systems. In: Proceedings of 18th International Conference on Spoken Language Translation (IWSLT) . Bangkok, on-line: Association for Computational Linguistics, 2021, pp. 75-83. ISBN 978-1-7138-3378-9. Available from: https://aclanthology.org/2021.iwslt-1.7.pdf
Czech title
BUT systém pro strojový překlad z řeči pro IWSLT 2021
Type
conference paper
Language
english
Authors
URL
Keywords

speech, translation

Abstract

The paper describes BUTs English to German offline speech translation (ST) systems developed for IWSLT2021. They are based on jointly trained Automatic Speech Recognition- Machine Translation models. Their performances is evaluated on MustC-Common test set. In this work, we study their efficiency from the perspective of having a large amount of separate ASR training data and MT training data, and a smaller amount of speechtranslation training data. Large amounts of ASR and MT training data are utilized for pretraining the ASR and MT models. Speechtranslation data is used to jointly optimize ASR-MT models by defining an end-to-end differentiable path from speech to translations. For this purpose, we use the internal continuous representations from the ASR-decoder as the input to MT module. We show that speech translation can be further improved by training the ASR-decoder jointly with the MT-module using large amount of text-only MT training data. We also show significant improvements by training an ASR module capable of generating punctuated text, rather than leaving the punctuation task to the MT module.

Published
2021
Pages
75-83
Proceedings
Proceedings of 18th International Conference on Spoken Language Translation (IWSLT)
Conference
18TH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE TRANSLATION, Bangkok (on-line), TH
ISBN
978-1-7138-3378-9
Publisher
Association for Computational Linguistics
Place
Bangkok, on-line, TH
DOI
UT WoS
000694723100007
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB12702,
   author = "K. Hari Vydana and Martin Karafi\'{a}t and Luk\'{a}\v{s} Burget and Jan \v{C}ernock\'{y}",
   title = "The IWSLT 2021 BUT Speech Translation Systems",
   pages = "75--83",
   booktitle = "Proceedings of 18th International Conference on Spoken Language Translation (IWSLT) ",
   year = 2021,
   location = "Bangkok, on-line, TH",
   publisher = "Association for Computational Linguistics",
   ISBN = "978-1-7138-3378-9",
   doi = "10.18653/v1/2021.iwslt-1.7",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12702"
}
Back to top