Publication detail

Unsupervised Word Segmentation from Speech with Attention

GODARD, P. BOITO, M. ONDEL YANG, L. BERARD, A. YVON, F. VILLAVICENCIO, A. BESACIER, L.

Original Title

Unsupervised Word Segmentation from Speech with Attention

Type

conference paper

Language

English

Original Abstract

We present a first attempt to perform attentional word segmentation directly from the speech signal, with the final goal to automatically identify lexical units in a low-resource, unwritten language (UL). Our methodology assumes a pairing between recordings in the UL with translations in a well-resourced language. It uses Acoustic Unit Discovery (AUD) to convert speech into a sequence of pseudo-phones that is segmented using neural soft-alignments produced by a neural machine translation model. Evaluation uses an actual Bantu UL, Mboshi; comparisons to monolingual and bilingual baselines illustrate the potential of attentional word segmentation for language documentation.

Keywords

computational language documentation,encoder-decoder models, attentional models, unsupervised word segmentation.

Authors

GODARD, P.; BOITO, M.; ONDEL YANG, L.; BERARD, A.; YVON, F.; VILLAVICENCIO, A.; BESACIER, L.

Released

2. 9. 2018

Publisher

International Speech Communication Association

Location

Hyderabad

ISBN

1990-9772

Periodical

Proceedings of Interspeech

Year of study

2018

Number

9

State

French Republic

Pages from

2678

Pages to

2682

Pages count

5

URL

https://www.isca-speech.org/archive/Interspeech_2018/pdfs/1308.pdf

BibTex

@inproceedings{BUT163406,
  author="GODARD, P. and BOITO, M. and ONDEL YANG, L. and BERARD, A. and YVON, F. and VILLAVICENCIO, A. and BESACIER, L.",
  title="Unsupervised Word Segmentation from Speech with Attention",
  booktitle="Proceeding of Interspeech 2018",
  year="2018",
  journal="Proceedings of Interspeech",
  volume="2018",
  number="9",
  pages="2678--2682",
  publisher="International Speech Communication Association",
  address="Hyderabad",
  doi="10.21437/Interspeech.2018-1308",
  issn="1990-9772",
  url="https://www.isca-speech.org/archive/Interspeech_2018/pdfs/1308.pdf"
}