HAL : hal-00663837, version 1

Fiche détaillée

Récupérer au format BibTeX - EndNote - TEI - RefWorks -

Speech Prosody, Shanghai : Chine (2012)

Making Sense of Variations: Introducing Alternatives in Speech Synthesis

Nicolas Obin 1, Christophe Veaux 1, 2, Pierre Lanchantin 1, 3

(25/05/2012)

This paper addresses the use of speech alternatives to enrich speech synthesis systems. Speech alternatives denote the variety of strategies that a speaker can use to pronounce a sentence - depending on pragmatic constraints, speaking style, and specific strategies of the speaker. During the training, symbolic and acoustic characteristics of a unit-selection speech synthesis system are statistically modelled with context-dependent parametric models (GMMs/HMMs). During the synthesis, symbolic and acoustic alternatives are exploited using a GENERALIZED VITERBI ALGORITHM (GVA) to determine the sequence of speech units used for the synthesis. Objective and subjective evaluations supports evidence that the use of speech alternatives significantly improves speech synthesis over conventional speech synthesis systems.

1 :	Sciences et Technologies de la Musique et du Son (STMS)
	IRCAM – CNRS : UMR9912 – Université Paris VI - Pierre et Marie Curie
2 :	Centre for Speech Technology Research (CSTR)
	University of Edinburgh
3 :	Cambridge University Engineering Department (CUED)
	University of Cambridge

Domaine

:

Sciences de l'ingénieur/Traitement du signal et de l'image Informatique/Traitement du signal et de l'image Statistiques/Applications Sciences de l'Homme et Société/Linguistique

Mots Clés : speech synthesis – speech prosody – speech alternatives

hal-00663837, version 1

http://hal.archives-ouvertes.fr/hal-00663837

oai:hal.archives-ouvertes.fr:hal-00663837

Contributeur : Nicolas Obin <>

Soumis le : Vendredi 27 Janvier 2012, 16:36:36

Dernière modification le : Vendredi 27 Janvier 2012, 16:44:55