Selective Tap Training of FIR filters for Blind Source Separation of Convolutive Speech Mixtures - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2009

Selective Tap Training of FIR filters for Blind Source Separation of Convolutive Speech Mixtures

Résumé

This paper presents a novel low complexity time domain algorithm for blind separation of speech signal from their convolutive mixtures. We try to reduce intrinsic computational complexity of time domain algorithms by adapting only a small subset of taps from separating FIR filters which are expected to attain largest values. This selection is accomplished by recovering spatial dependencies using linear prediction (LP) analysis. Then we use particle swarm optimization (PSO) in order to find best values for these selected taps. We employ the sparseness properties of speech signals in the time-frequency (TF) domain to define a low complexity and yet appropriate fitness function which numerically quantifies the amount of achieved separation by each one of the particles during PSO execution.
Fichier non déposé

Dates et versions

inria-00438670 , version 1 (04-12-2009)

Identifiants

  • HAL Id : inria-00438670 , version 1

Citer

Ali Khanagha, Vahid Khanagha, Mohamad Hussein Kahaei. Selective Tap Training of FIR filters for Blind Source Separation of Convolutive Speech Mixtures. 2009 IEEE Symposium on Industrial Electronics and Applications., Oct 2009, Kuala Lumpur, Malaysia. ⟨inria-00438670⟩

Collections

INRIA INRIA2
80 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More