An investigation of likelihood normalization for robust ASR

Abstract : Noise-robust automatic speech recognition (ASR) systems rely on feature and/or model compensation. Existing compensation techniques typically operate on the features or on the parameters of the acoustic models themselves. By contrast, a number of normalization techniques have been defined in the field of speaker verification that operate on the resulting log-likelihood scores. In this paper, we provide a theoretical motivation for likelihood normalization due to the so-called "hubness" phenomenon and we evaluate the benefit of several normalization techniques on ASR accuracy for the 2nd CHiME Challenge task. We show that symmetric normalization (S-norm) reduces the relative error rate by 43% alone and by 10% after feature and model compensation.
Type de document :
Communication dans un congrès
Interspeech, Sep 2014, Singapore, Singapore. 2014
Liste complète des métadonnées

Littérature citée [35 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01006142
Contributeur : Emmanuel Vincent <>
Soumis le : vendredi 13 juin 2014 - 21:17:46
Dernière modification le : jeudi 11 janvier 2018 - 06:25:24
Document(s) archivé(s) le : samedi 13 septembre 2014 - 11:40:29

Fichier

vincent_IS14.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01006142, version 1

Citation

Emmanuel Vincent, Aggelos Gkiokas, Dominik Schnitzer, Arthur Flexer. An investigation of likelihood normalization for robust ASR. Interspeech, Sep 2014, Singapore, Singapore. 2014. 〈hal-01006142〉

Partager

Métriques

Consultations de la notice

399

Téléchargements de fichiers

501