Accéder directement au contenu Accéder directement à la navigation
Pré-publication, Document de travail

User-friendly automatic transcription of low-resource languages: Plugging ESPnet into Elpis

Oliver Adams 1 Benjamin Galliot 2 Guillaume Wisniewski 3 Nicholas Lambourne 4 Ben Foley 4, 5 Rahasya Sanders-Dwyer 4, 5 Janet Wiles 4, 5 Alexis Michaud 2 Séverine Guillaume 2 Laurent Besacier 6 Christopher Cox 7 Katya Aplonova 8 Guillaume Jacques 9 Nathan Hill 10
1 Atos zData
2 LACITO - Langues et civilisations à tradition orale
3 LLF UMR7110 - Laboratoire de Linguistique Formelle
4 University of Queensland [Brisbane]
5 ARC Centre of Excellence for the Dynamics of Language (CoEDL)
6 LIG - Laboratoire d'Informatique de Grenoble
7 University of Alberta
8 LLACAN - Langage, LAngues et Cultures d'Afrique
9 CRLAO - Centre de Recherches Linguistiques sur l'Asie Orientale
10 School of Oriental and African Studies (SOAS), University of London
Abstract : This paper reports on progress integrating the speech recognition toolkit ESPnet into Elpis, a web front-end originally designed to provide access to the Kaldi automatic speech recognition toolkit. The goal of this work is to make end-to-end speech recognition models available to language workers via a user-friendly graphical interface. Encouraging results are reported on (i) development of an ESPnet recipe for use in Elpis, with preliminary results on data sets previously used for training acoustic models with the Persephone toolkit along with a new data set that had not previously been used in speech recognition, and (ii) incorporating ESPnet into Elpis along with UI enhancements and a CUDA-supported Dockerfile.
Keywords : automatic transcription automatic speech recognition language documentation endangered languages Computational Language Documentation
Type de document :
Pré-publication, Document de travail
Liste complète des métadonnées
https://halshs.archives-ouvertes.fr/halshs-03030529
Contributeur : Alexis Michaud <>
Soumis le : lundi 14 décembre 2020 - 14:04:27
Dernière modification le : jeudi 15 avril 2021 - 15:08:16
Archivage à long terme le : : lundi 15 mars 2021 - 18:04:57

Fichiers

Vignette du document
Elpis_ComputEL4.pdf
Fichiers produits par l'(les) auteur(s)

Licence

Distributed under a Creative Commons Paternité - Pas d'utilisation commerciale - Partage selon les Conditions Initiales 4.0 International License

Identifiants

  • HAL Id : halshs-03030529, version 1
  • ARXIV : 2101.03027

Citation

Oliver Adams, Benjamin Galliot, Guillaume Wisniewski, Nicholas Lambourne, Ben Foley, et al.. User-friendly automatic transcription of low-resource languages: Plugging ESPnet into Elpis. 2020. ⟨halshs-03030529v1⟩

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

242

Téléchargements de fichiers

167