Fully Automated Non-Native Speech Recognition Using Confusion-Based Acoustic Model Integration And Graphemic Constraints

Ghazi Bouselmi 1 Dominique Fohr 1 Irina Illina 1 Jean-Paul Haton 1
1 PAROLE - Analysis, perception and recognition of speech
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : This paper presents a fully automated approach for the recognition of non-native speech based on acoustic model modification. For a native language (L1) and a spoken language (L2), pronunciation variants of the phones of L2 are automatically extracted from an existing non-native database as a confusion matrix with sequences of phones of L1. This is done using L1's and L2's ASR systems. This confusion concept deals with the problem of non existence of match between some L2 and L1 phones. The confusion matrix is then used to modify the acoustic models (HMMs) of L2 phones by integrating corresponding L1 phone models as alternative HMM paths. We introduce graphemic contraints in the confusion extraction process: the phonetic confusion is established for each couple of `L2-phone' and the grapheme(s) correspondig to that phone. We claim that prononciation errors may depend on the graphemes related to each phone. The modified ASR system achieved an improvement between 32% and 40% (relative, L1=French and L2=English) in WER on the French non-native database used for testing. The introduction of graphemic contraints in the phonetic confusion allowed further improvements.
keyword : non-native HMM model modification pronunciation modelling
Type de document :
Communication dans un congrès
IEEE International Conference on Acoustics, Speech, and Signal Processing - ICASSP 2006, May 2006, Toulouse/France, 2006
Liste complète des métadonnées

Littérature citée [7 références]  Voir  Masquer  Télécharger
https://hal.inria.fr/inria-00110492
Contributeur : Bouselmi Ghazi <>
Soumis le : samedi 9 décembre 2006 - 20:04:21
Dernière modification le : jeudi 11 janvier 2018 - 06:19:56
Document(s) archivé(s) le : mardi 6 avril 2010 - 21:15:41

Fichier

icassp-2006.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00110492, version 1

Collections

INRIA | LORIA | UNIV-LORRAINE

Citation

Ghazi Bouselmi, Dominique Fohr, Irina Illina, Jean-Paul Haton. Fully Automated Non-Native Speech Recognition Using Confusion-Based Acoustic Model Integration And Graphemic Constraints. IEEE International Conference on Acoustics, Speech, and Signal Processing - ICASSP 2006, May 2006, Toulouse/France, 2006. 〈inria-00110492〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

331

Téléchargements de fichiers

175

Contact


archive-ouverte@inria.fr