Publication Details

Unsupervised Word Segmentation from Discrete Speech Units in Low-Resource Settings

BOITO, M.; YUSUF, B.; ONDEL YANG, L.; VILLAVICENCIO, A.; BESACIER, L. Unsupervised Word Segmentation from Discrete Speech Units in Low-Resource Settings. In Proceedings of the the 1st Annual Meeting of the ELRA/ISCA Special Interest Group on Under-Resourced Languages. Marseile: European Language Resources Association, 2022. p. 1-9. ISBN: 979-10-95546-91-7.
Czech title
Slovní segmentace s učením dohledu založená na diskrétních řečových jednotkách ve scénářích s nedostatkem zdrojů
Type
conference paper
Language
English
Authors
BOITO, M.
Yusuf Bolaji (DCGM)
ONDEL YANG, L.
VILLAVICENCIO, A.
BESACIER, L.
URL
Keywords

unsupervised word segmentation, speech discretization, acoustic unit discovery, low-resource settings

Abstract

Documenting languages helps to prevent the extinction of endangered dialects - many of which are otherwise expected to dis- appear by the end of the century. When documenting oral languages, unsupervised word segmentation (UWS) from speech is a useful, yet challenging, task. It consists in producing time-stamps for slicing utterances into smaller segments corresponding to words, being performed from phonetic transcriptions, or in the absence of these, from the output of unsupervised speech discretization models. These discretization models are trained using raw speech only, producing discrete speech units that can be applied for downstream (text-based) tasks. In this paper we compare five of these models: three Bayesian and two neural approaches, with regards to the exploitability of the produced units for UWS. For the UWS task, we experiment with two models, using as our target language the Mboshi (Bantu C25), an unwritten language from Congo-Brazzaville. Additionally, we report results for Finnish, Hungarian, Romanian and Russian in equally low-resource settings, using only 4 hours of speech. Our results suggest that neural models for speech discretization are difficult to exploit in our setting, and that it might be necessary to adapt them to limit sequence length. We obtain our best UWS results by using Bayesian models that produce high quality, yet compressed, discrete representations of the input speech signal.

Published
2022
Pages
1–9
Proceedings
Proceedings of the the 1st Annual Meeting of the ELRA/ISCA Special Interest Group on Under-Resourced Languages
ISBN
979-10-95546-91-7
Publisher
European Language Resources Association
Place
Marseile
EID Scopus
2-s2.0-85146249954
BibTeX 
@inproceedings{BUT187752,
  author="BOITO, M. and YUSUF, B. and ONDEL YANG, L. and VILLAVICENCIO, A. and BESACIER, L.",
  title="Unsupervised Word Segmentation from Discrete Speech Units in Low-Resource Settings",
  booktitle="Proceedings of the the 1st Annual Meeting of the ELRA/ISCA Special Interest Group on Under-Resourced Languages",
  year="2022",
  pages="1--9",
  publisher="European Language Resources Association",
  address="Marseile",
  isbn="979-10-95546-91-7",
  url="https://aclanthology.org/2022.sigul-1.1.pdf"
}
Files
Back to top