Publication Details

BUT Systems for IWSLT 2023 Marathi - Hindi Low Resource Speech Translation Task

KESIRAJU, S.; BENEŠ, K.; TIKHONOV, M.; ČERNOCKÝ, J. BUT Systems for IWSLT 2023 Marathi - Hindi Low Resource Speech Translation Task. In 20th International Conference on Spoken Language Translation, IWSLT 2023 - Proceedings of the Conference. Toronto (in-person and online): Association for Computational Linguistics, 2023. p. 227-234. ISBN: 978-1-959429-84-5.
Czech title
BUT Systémy pro IWSLT 2023 úlohu překladu řeči z maráthštiny do hindštiny
Type
conference paper
Language
English
Authors
URL
Keywords

Marathi, Hindi, Low Resource, Speech, Translation

Abstract

This paper describes the systems submitted for Marathi to Hindi low-resource
speech translation task. Our primary submission is based on an end-to-end direct
speech translation system, whereas the contrastive one is a cascaded system. The
backbone of both the systems is a Hindi-Marathi bilingual ASR system trained on
2790 hours of imperfect transcribed speech. The end-to-end speech translation
system was directly initialized from the ASR, and then finetuned for direct
speech translation with an auxiliary CTC loss for translation. The MT model for
the cascaded system is initialized from a cross-lingual language model, which was
then fine-tuned using 1.6 M parallel sentences. All our systems were trained from
scratch on publicly available datasets. In the end, we use a language model to
re-score the n-best hypotheses. Our primary submission achieved 30.5 and 39.6
BLEU whereas the contrastive system obtained 21.7 and 28.6 BLEU on official dev
and test sets respectively. The paper also presents the analysis on several
experiments that were conducted and outlines the strategies for improving speech
translation in low-resource scenarios.

Published
2023
Pages
227–234
Proceedings
20th International Conference on Spoken Language Translation, IWSLT 2023 - Proceedings of the Conference
Conference
The 61st Annual Meeting of the Association for Computational Linguistics, Toronto, CA
ISBN
978-1-959429-84-5
Publisher
Association for Computational Linguistics
Place
Toronto (in-person and online)
DOI
EID Scopus
BibTeX
@inproceedings{BUT185198,
  author="Santosh {Kesiraju} and Karel {Beneš} and Maksim {Tikhonov} and Jan {Černocký}",
  title="BUT Systems for IWSLT 2023 Marathi - Hindi Low Resource Speech Translation Task",
  booktitle="20th International Conference on Spoken Language Translation, IWSLT 2023 -  Proceedings of the Conference",
  year="2023",
  pages="227--234",
  publisher="Association for Computational Linguistics",
  address="Toronto (in-person and online)",
  doi="10.18653/v1/2023.iwslt-1.19",
  isbn="978-1-959429-84-5",
  url="https://aclanthology.org/2023.iwslt-1.19.pdf"
}
Files
Back to top