Publication result detail

Speech Technology for Unwritten Languages

SCHARENBORG, O.; BESACIER, L.; BLACK, A.; HASEGAWA-JOHNSON, M.; METZE, F.; NEUBIG, G.; STÜKER, S.; GODARD, P.; MÜLLER, M.; ONDEL YANG, L.; PALASKAR, S.; ARTHUR, P.; CIANNELLA, F.; DU, M.; LARSEN, E.; MERKX, D.; RIAD, R.; WANG, L.; DUPOUX, E.

Original Title

Speech Technology for Unwritten Languages

English Title

Speech Technology for Unwritten Languages

Type

WoS Article

Original Abstract

Abstract-Speech technology plays an important role in oureveryday life. Among others, speech is used for human-computerinteraction, for instance for information retrieval and on-lineshopping. In the case of an unwritten language, however, speechtechnology is unfortunately difficult to create, because it cannotbe created by the standard combination of pre-trained speech-to-text and text-to-speech subsystems. The research presented in this article takes the first steps towards speech technology forunwritten languages. Specifically, the aim of this work was 1) tolearn speech-to-meaning representations without using text as anintermediate representation, and 2) to test the sufficiency of thelearned representations to regenerate speech or translated text, orto retrieve images that depict the meaning of an utterance in anunwritten language. The results suggest that building systems thatgo directly from speech-to-meaning and from meaning-to-speech,bypassing the need for text, is possible.

English abstract

Abstract-Speech technology plays an important role in oureveryday life. Among others, speech is used for human-computerinteraction, for instance for information retrieval and on-lineshopping. In the case of an unwritten language, however, speechtechnology is unfortunately difficult to create, because it cannotbe created by the standard combination of pre-trained speech-to-text and text-to-speech subsystems. The research presented in this article takes the first steps towards speech technology forunwritten languages. Specifically, the aim of this work was 1) tolearn speech-to-meaning representations without using text as anintermediate representation, and 2) to test the sufficiency of thelearned representations to regenerate speech or translated text, orto retrieve images that depict the meaning of an utterance in anunwritten language. The results suggest that building systems thatgo directly from speech-to-meaning and from meaning-to-speech,bypassing the need for text, is possible.

Keywords

Speech processing, automatic speech recognition,unsupervised learning, speech synthesis, image retrieval.

Key words in English

Speech processing, automatic speech recognition,unsupervised learning, speech synthesis, image retrieval.

Authors

SCHARENBORG, O.; BESACIER, L.; BLACK, A.; HASEGAWA-JOHNSON, M.; METZE, F.; NEUBIG, G.; STÜKER, S.; GODARD, P.; MÜLLER, M.; ONDEL YANG, L.; PALASKAR, S.; ARTHUR, P.; CIANNELLA, F.; DU, M.; LARSEN, E.; MERKX, D.; RIAD, R.; WANG, L.; DUPOUX, E.

RIV year

2021

Released

13.02.2020

ISBN

2329-9290

Periodical

IEEE-ACM Transactions on Audio Speech and Language Processing

Volume

2020

Number

28

State

United States of America

Pages from

964

Pages to

975

Pages count

12

URL

https://ieeexplore.ieee.org/document/8998182

BibTex

@article{BUT170325,
  author="SCHARENBORG, O. and BESACIER, L. and BLACK, A. and HASEGAWA-JOHNSON, M. and METZE, F. and NEUBIG, G. and STÜKER, S. and GODARD, P. and MÜLLER, M. and ONDEL YANG, L. and PALASKAR, S. and ARTHUR, P. and CIANNELLA, F. and DU, M. and LARSEN, E. and MERKX, D. and RIAD, R. and WANG, L. and DUPOUX, E.",
  title="Speech Technology for Unwritten Languages",
  journal="IEEE-ACM Transactions on Audio Speech and Language Processing",
  year="2020",
  volume="2020",
  number="28",
  pages="964--975",
  doi="10.1109/TASLP.2020.2973896",
  issn="2329-9290",
  url="https://ieeexplore.ieee.org/document/8998182"
}

Documents

scharenborg_ieee_ACM_transactions2020_08998182