Nonnegative matrix factorization and spatial covariance model for under-determined reverberant audio source separation - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2010

Nonnegative matrix factorization and spatial covariance model for under-determined reverberant audio source separation

Simon Arberet
  • Function : Author
  • PersonId : 882921
Alexey Ozerov
  • Function : Author
  • PersonId : 882775
Pierre Vandergheynst
  • Function : Author
  • PersonId : 839985

Abstract

We address the problem of blind audio source separation in the under-determined and convolutive case. The contribution of each source to the mixture channels in the time-frequency domain is modeled by a zero-mean Gaussian random vector with a full rank covariance matrix composed of two terms: a variance which represents the spectral properties of the source and which is modeled by a nonnegative matrix factorization (NMF) model and another full rank covariance matrix which encodes the spatial properties of the source contribution in the mixture. We address the estimation of these parameters by maximizing the likelihood of the mixture using an expectation-maximization (EM) algorithm. Theoretical propositions are corroborated by experimental studies on stereo reverberant music mixtures.
Fichier principal
Vignette du fichier
2010_ISSPA_ArberetEtAl.pdf (182.83 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

inria-00541436 , version 1 (05-02-2011)

Identifiers

Cite

Simon Arberet, Alexey Ozerov, Ngoc Q. K. Duong, Emmanuel Vincent, Rémi Gribonval, et al.. Nonnegative matrix factorization and spatial covariance model for under-determined reverberant audio source separation. Information Sciences Signal Processing and their Applications (ISSPA), 2010 10th International Conference on, May 2010, Kuala Lumpur, Malaysia. pp.1--4, ⟨10.1109/ISSPA.2010.5605570⟩. ⟨inria-00541436⟩
602 View
781 Download

Altmetric

Share

Gmail Facebook X LinkedIn More