Amharic Speech Recognition for Speech Translation

Abstract : The state-of-the-art speech translation can be seen as a cascade of Automatic Speech Recognition, Statistical Machine Translation and Text-To-Speech synthesis. In this study an attempt is made to experiment on Amharic speech recognition for Amharic-English speech translation in tourism domain. Since there is no Amharic speech corpus, we developed a read-speech corpus of 7.43hr in tourism domain. The Amharic speech corpus has been recorded after translating standard Basic Traveler Expression Corpus (BTEC) under a normal working environment. In our ASR experiments phoneme and syllable units are used for acoustic models, while morpheme and word are used for language models. Encouraging ASR results are achieved using morpheme-based language models and phoneme-based acoustic models with a recognition accuracy result of 89.1%, 80.9%, 80.6%, and 49.3% at character, morph, word and sentence level respectively. We are now working towards designing Amharic-English speech translation through cascading components under different error correction algorithms.
Type de document :
Communication dans un congrès
Atelier Traitement Automatique des Langues Africaines (TALAF). JEP-TALN 2016, Jul 2016, Paris, France. Actes de la conférence conjointe JEP-TALN-RECITAL 2016
Liste complète des métadonnées

Littérature citée [14 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01350050
Contributeur : Laurent Besacier <>
Soumis le : vendredi 29 juillet 2016 - 15:12:01
Dernière modification le : jeudi 11 octobre 2018 - 08:48:03
Document(s) archivé(s) le : dimanche 30 octobre 2016 - 11:30:28

Fichier

WOLDEYOHANNIS_ET_AL - Amharic ...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01350050, version 1

Collections

Citation

Michael Melese, Laurent Besacier, Million Meshesha. Amharic Speech Recognition for Speech Translation. Atelier Traitement Automatique des Langues Africaines (TALAF). JEP-TALN 2016, Jul 2016, Paris, France. Actes de la conférence conjointe JEP-TALN-RECITAL 2016. 〈hal-01350050〉

Partager

Métriques

Consultations de la notice

183

Téléchargements de fichiers

910