Skip to Main content Skip to Navigation
New interface
Conference papers

Séparation de sources à partir d'un seul capteur pour la reconnaissance robuste de la parole

Guillaume Gravier 1 Laurent Benaroya 1 Alexey Ozerov 1 Rémi Gribonval 1 Frédéric Bimbot 1 
1 METISS - Speech and sound data modeling and processing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : In this paper, we address the problem of noise compensation in speech signals for robust speech recognition. Several classical denoising methods in the field of speech and signal processing are compared on speech corrupted by music, as it is often the case in broadcast news transcription tasks. We also present two new source separation techniques, namely adaptive Wiener filtering and adaptive shrinkage. These techniques rely on the use of a dictionary of spectral shapes in order to tackle the problem of non stationarity of the signals. The algorithms are first compared on the source separation task and assessed in terms of average distortion. Their effect on the entire transcription system is eventually compared in terms of word error rate. Results show that the proposed adaptive Wiener filter approach yields a significant improvement of the transcription accuracy at signal/noise ratios greater than 15 dB.
Complete list of metadata

Cited literature [10 references]  Display  Hide  Download
Contributor : Rémi Gribonval Connect in order to contact the contributor
Submitted on : Sunday, February 20, 2011 - 9:23:21 PM
Last modification on : Friday, February 4, 2022 - 3:19:41 AM
Long-term archiving on: : Saturday, May 21, 2011 - 2:43:51 AM


Files produced by the author(s)


  • HAL Id : inria-00567339, version 1


Guillaume Gravier, Laurent Benaroya, Alexey Ozerov, Rémi Gribonval, Frédéric Bimbot. Séparation de sources à partir d'un seul capteur pour la reconnaissance robuste de la parole. Journées d'Etude sur la Parole: JEP 2004, Apr 2004, Fès, Maroc. ⟨inria-00567339⟩



Record views


Files downloads