Exploiting models intrinsic robustness for noisy speech recognition - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2004

Exploiting models intrinsic robustness for noisy speech recognition

Résumé

We propose in this paper an original approach to build masks in the framework of missing data recognition. The proposed soft masks are estimated from the models themselves, and not from the test signal as it is usually the case. They represent the intrinsic robustness of model's log-spectral coefficients. The method is validated with cepstral models, on two synthetic and two real-life noises, at different signal-to-noise ratios. We further discuss how such masks can be combined with other signal-based masks and noise compensation techniques.
Fichier principal
Vignette du fichier
A04-R-270.pdf (141.28 Ko) Télécharger le fichier
Loading...

Dates et versions

inria-00099890 , version 1 (26-09-2006)

Identifiants

  • HAL Id : inria-00099890 , version 1

Citer

Christophe Cerisara, Dominique Fohr, Odile Mella, Irina Illina. Exploiting models intrinsic robustness for noisy speech recognition. 8th International Conference on Spoken Language Processing - ICSLP'2004, 2004, Jeju, Corée du Sud, 4 p. ⟨inria-00099890⟩
102 Consultations
71 Téléchargements

Partager

Gmail Facebook X LinkedIn More