A corpus-based learning method for prominence detection in spontaneous speech - HAL Accéder directement au contenu
Communication dans un congrès Année : 2010

A corpus-based learning method for prominence detection in spontaneous speech

Résumé

The aim of this paper is to present a software tool called ANALOR, which allows semi-automatic prominence detection in spontaneous French. On the basis of a manual annotation performed by two experts on a 70-minute long corpus including different regional varieties of French (Belgian, Swiss and metropolitan French) and various discourse genres (from read speech to spontaneous conversations), our system conducts a learning-method in order to determine the best thresholds for prominence prediction. This procedure appreciably improves detection, with consistency between automatic identification and the human labeling rising from 75.3 without training to 79.1 of f-measure after corpus-based learning.

Mots clés

Fichier principal
Vignette du fichier
17.Speech-Prosody.pdf ( 94.9 Ko ) Télécharger
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

halshs-00637632, version 1 (02-11-2011)

Identifiants

  • HAL Id : halshs-00637632 , version 1

Citer

Anne Lacheret, Mathieu Avanzi, Bernard Victorri. A corpus-based learning method for prominence detection in spontaneous speech. Speech Prosody, 2010, Chicago, United States. pp.20-30. ⟨halshs-00637632⟩
188 Consultations
178 Téléchargements
Dernière date de mise à jour le 28/04/2024
comment ces indicateurs sont-ils produits

Partager

Gmail Facebook Twitter LinkedIn Plus