Combined systems for automatic phonetic transcription of proper nouns

Abstract : Large vocabulary automatic speech recognition (ASR) technologies perform well in known, controlled contexts. However recognition of proper nouns is commonly considered as a difficult task. Accurate phonetic transcription of a proper noun is difficult to obtain, although it can be one of the most important resources for a recognition system. In this article, we propose methods of automatic phonetic transcription applied to proper nouns. The methods are based on combinations of the rule-based phonetic transcription generator LIA PHON and an acoustic-phonetic decoding system. On the ESTER corpus, we observed that the combined systems obtain better results than our reference system (LIA PHON). The WER (Word Error Rate) decreased on segments of speech containing proper nouns, without affecting negatively the results on the rest of the corpus. On the same corpus, the Proper Noun Error Rate (PNER, which is a WER computed on proper nouns only), decreased with our new system.
Type de document :
Communication dans un congrès
LREC-2008, May 2008, Marrakech, Morocco. LREC-2008 proceedings, 〈http://www.lrec-conf.org/lrec2008/〉
Liste complète des métadonnées

Littérature citée [8 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01502832
Contributeur : Antoine Laurent <>
Soumis le : jeudi 6 avril 2017 - 11:32:26
Dernière modification le : mardi 12 septembre 2017 - 12:10:37
Document(s) archivé(s) le : vendredi 7 juillet 2017 - 13:01:09

Fichier

Laurent08.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01502832, version 1

Collections

Citation

Antoine Laurent, Teva Merlin, Sylvain Meignier, Yannick Estève, Paul Deléglise. Combined systems for automatic phonetic transcription of proper nouns. LREC-2008, May 2008, Marrakech, Morocco. LREC-2008 proceedings, 〈http://www.lrec-conf.org/lrec2008/〉. 〈hal-01502832〉

Partager

Métriques

Consultations de la notice

110

Téléchargements de fichiers

45