On transforming spectral peaks in voice conversion - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

On transforming spectral peaks in voice conversion

Résumé

This paper explores the benefits of transforming spectral peaks in voice conversion. First, in examining classic GMMbased transformation with cepstral coefficients, we show that the lack of transformed data variance ("over-smoothing") can be related to the choice of spectral parameterization. Consequently, we propose an alternative parameterization using spectral peaks. The peaks are transformed using HMMs with Gaussian state distributions. Two learning variants and post-processing treating peak evolution in time are also examined. In comparing the different transformation approaches, spectral peaks are shown to offer higher interspeaker feature correlation and yield higher transformed data variance than their cepstral coefficient counterparts.
Fichier principal
Vignette du fichier
SSW2010.pdf (63.35 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00725662 , version 1 (27-08-2012)

Identifiants

  • HAL Id : hal-00725662 , version 1

Citer

Elizabeth Godoy, Olivier Rosec, Thierry Chonavel. On transforming spectral peaks in voice conversion. SSW7: 7th ISCA Speech Synthesis Workshop, Sep 2010, Kyoto, Japan. ⟨hal-00725662⟩
169 Consultations
314 Téléchargements

Partager

Gmail Facebook X LinkedIn More