Under-determined source separation via mixed-norm regularized minimization - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2008

Under-determined source separation via mixed-norm regularized minimization

Résumé

We consider the problem of extracting the source signals from an under-determined convolutive mixture assuming that the mixing filters are known. We wish to exploit the sparsity and approximate disjointness of the time-frequency representations of the sources. However, classical time-frequency masking techniques cannot be directly applied due to the convolutive nature of the mixture. To address this problem, we first formulate it as the minimization of a functional combining a classical 2 discrepancy term between the observed mixture and the mixture reconstructed from the estimated sources and a sparse regularization term defined in terms of mixed 2 / 1 norms of source coefficients in a time-frequency domain. The minimum of the functional is then obtained by a thresholded Landweber iteration algorithm. Preliminary results are discussed for two synthetic audio mixtures.
Fichier principal
Vignette du fichier
Eusipco08_KVG.pdf (173.97 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00347089 , version 1 (14-12-2008)

Identifiants

  • HAL Id : hal-00347089 , version 1

Citer

Matthieu Kowalski, Emmanuel Vincent, Rémi Gribonval. Under-determined source separation via mixed-norm regularized minimization. European Signal Processing Conference 2008, Aug 2008, Lausanne, Switzerland. pp.1569103109. ⟨hal-00347089⟩
557 Consultations
165 Téléchargements

Partager

Gmail Facebook X LinkedIn More