Learning vocal tract variables with multi-task kernels - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2011

Learning vocal tract variables with multi-task kernels

Résumé

The problem of acoustic-to-articulatory speech inversion continues to be a challenging research problem which sig- nificantly impacts automatic speech recognition robustness and accuracy. This paper presents a multi-task kernel based method aimed at learning Vocal Tract (VT) variables from the Mel-Frequency Cepstral Coefficients (MFCCs). Unlike usual speech inversion techniques based on individual esti- mation of each tract variable, the key idea here is to consider all the target variables simultaneously to take advantage of the relationships among them and then improve learning per- formance. The proposed method is evaluated using synthetic speech dataset and corresponding tract variables created by the TAsk Dynamics Application (TADA) model and com- pared to the hierarchical ε-SVR speech inversion technique.
Fichier principal
Vignette du fichier
ICASSP2011.pdf (94.49 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00826050 , version 1 (04-06-2013)

Identifiants

  • HAL Id : hal-00826050 , version 1

Citer

Hachem Kadri, Emmanuel Duflos, Philippe Preux. Learning vocal tract variables with multi-task kernels. 36th International Conference on Acoustics, Speech and Signal Processing (ICASSP), May 2011, Prague, Czech Republic. ⟨hal-00826050⟩
214 Consultations
199 Téléchargements

Partager

Gmail Facebook X LinkedIn More