Fiche du document
2 mars 2021
- handle: 10670/1.u0k02h
- halshs: halshs-03030529
- ARXIV: 2101.03027
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/arxiv/2101.03027
Ce document est lié à :
info:eu-repo/grantAgreement/EC/FP7/609823/EU/Beyond Boundaries: Religion, Region, Language and the State/ASIA
Ce document est lié à :
info:eu-repo/grantAgreement//758232/EU/Discourse reporting in African storytelling/
http://creativecommons.org/licenses/by-nc-sa/ , info:eu-repo/semantics/OpenAccess
Mots-clés
automatic transcription language documentation endangered languages automatic speech recognition Computational Language Documentation MESH: reconnaissance automatique de la parole MESH: transcription automatique MESH: Traitement Automatique des Langues Naturelles MESH: documentation linguistique MESH: langues en danger ACM: D.: SoftwareSujets proches
Language (New words, slang, etc.)Citer ce document
Oliver Adams et al., « User-friendly automatic transcription of low-resource languages: Plugging ESPnet into Elpis », HAL-SHS : linguistique, ID : 10670/1.u0k02h
Métriques
Partage / Export
Résumé
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.