When mismatched training data outperform matched data

Emmanuel Vincent 1
1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : My talk will focus on robustness to background noise in distant-microphone speech recordings. I will introduce deep learning based techniques for speech enhancement and for acoustic modeling of speech. I will then report the results of a study on the impact of environment and microphone mismatches on the recognition accuracy. This study reveals that mismatched training data can sometimes outperform matched data. I will suggest a way to optimize the training set in order to exploit this finding.
Type de document :
Communication dans un congrès
Systematic approaches to deep learning methods for audio, Sep 2017, Vienna, Austria. 〈http://www.univie.ac.at/nuhag-php/event_NEW/make.php?event=esi17〉
Liste complète des métadonnées

Littérature citée [4 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01588876
Contributeur : Emmanuel Vincent <>
Soumis le : dimanche 17 septembre 2017 - 20:56:33
Dernière modification le : jeudi 11 janvier 2018 - 06:27:31
Document(s) archivé(s) le : lundi 18 décembre 2017 - 13:06:08

Fichier

vincent_ESI17.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01588876, version 1

Citation

Emmanuel Vincent. When mismatched training data outperform matched data. Systematic approaches to deep learning methods for audio, Sep 2017, Vienna, Austria. 〈http://www.univie.ac.at/nuhag-php/event_NEW/make.php?event=esi17〉. 〈hal-01588876〉

Partager

Métriques

Consultations de la notice

370

Téléchargements de fichiers

107