Accueil > Équipes > S2I > Matthieu Puigt > Matthieu Puigt ’s PhD thesis defense

 

Matthieu Puigt ’s PhD thesis defense

Par Matthieu Puigt - 27/04/2009

 

Blind source separation methods based on time-frequency transforms.
Application to speech signals


Committee

  • Pierre Comon, I3S, CNRS, Sophia-Antipolis, president
  • Ali Mansour, E3I2, ENSIETA, Brest, reviewer
  • Eric Moreau, MS, Université de Toulon-ISITV, reviewer
  • Jean-Philippe Bernard, CESR, CNRS, Toulouse, member
  • Shahram Hosseini, LATT, Université de Toulouse, member
  • Yannick Deville, LATT, Université de Toulouse, thesis supervisor

  • Thesis abstract

    Several time-frequency (TF) blind source separation (BSS) methods have been proposed in this thesis. In the systems output that have been used, a contribution of each source is estimated, using only mixed signals. All the methods proposed in this manuscript find tiny TF zones where only one source is active and estimate the mixing parameters in these zones. These approaches are particularly well suited for non-stationary sources (speech, music).

    We first studied and improved linear instantaneous methods based on variance or correlation criteria, that have been previously proposed by our team. They yield excellent performance for speech signals and can also separate spectra from astrophysical data. However, the nature of the mixtures that they can process limits their application fields.

    We have extended these approaches to more realistic mixtures. The first extensions consider attenuated and delayed mixtures of sources, which corresponds to mixtures in anechoic chamber. They require less restrictive sparsity assumptions than some approaches previously proposed in the literature, while addressing the same type of mixtures. We have studied the contribution of clustering techniques to our approaches and have achieved good performance for mixtures of speech signals.

    Lastly, a theoretical extension of these methods to general convolutive mixtures is described. It needs strong sparsity hypotheses and we have to solve classical indeterminacies of frequency-domain BSS methods.

    Keywords : blind source separation, linear instantaneous mixtures, attenuated and delayed mixtures, convolutive mixtures, time-frequecy analysis, non-stationnary sources, short-time Fourier transform, sparsity, correlation, variance, clustering, speech, astrophysics.


    Practical informations

  • Date : December 13, 2007
  • Hour : 11h00
  • Place : Conference room, CESR, 9 Av du Colonel Roche, 31028 Toulouse, France
  •  

     

    [Page Précédente] [Dans la même rubrique] [Sommaire]

     

     

    Site du CNRS Logo OMP
    Logo UPS
    Logo INSU-CNRS

    Annuaire

    Rechercher

    Sur ce site



    Bibliotheque

    Laboratoire Astrophysique de Toulouse - Tarbes (UMR5572)

    CNRS (Midi-Pyrénées)

    Univ. Paul Sabatier

    Liens utiles

    Accueil Imprimer Contact mail Plan du site Crédits