Hybrid model and structured sparsity for under-determined convolutive audio source separation - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Hybrid model and structured sparsity for under-determined convolutive audio source separation

Fangchen Feng
  • Fonction : Auteur
  • PersonId : 957219
Matthieu Kowalski

Résumé

We consider the problem of extracting the source signals from an under-determined convolutive mixture, assuming known filters. We start from its formulation as a minimization of a convex functional, combining a classical $\ell_2$ discrepancy term between the observed mixture and the one reconstructed from the estimated sources, and a sparse regularization term of source coefficients in a time-frequency domain. We then introduce a first kind of structure, using a hybrid model. Finally, we embed the previously introduced Windowed-Group-Lasso operator into the iterative thresholding/shrinkage algorithm, in order to take into account some structures inside each layers of time-frequency representations. Intensive numerical studies confirm the benefits of such an approach.
Fichier principal
Vignette du fichier
FK_icassp14.pdf (115.88 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01002996 , version 1 (10-06-2014)

Identifiants

Citer

Fangchen Feng, Matthieu Kowalski. Hybrid model and structured sparsity for under-determined convolutive audio source separation. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2014), May 2014, Florence, Italy. pp.AASP-P9.9, ⟨10.1109/icassp.2014.6854893⟩. ⟨hal-01002996⟩
160 Consultations
256 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More