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

Abstract : 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.
Type de document :
Communication dans un congrès
2009 IEEE Symposium on Industrial Electronics and Applications (ISIEA 2009), Oct 2009, Kuala Lumpur, Malaysia. 2009
Liste complète des métadonnées

Littérature citée [7 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00938354
Contributeur : H. Yahia <>
Soumis le : mercredi 29 janvier 2014 - 12:35:16
Dernière modification le : mercredi 29 novembre 2017 - 09:22:57
Document(s) archivé(s) le : lundi 5 mai 2014 - 11:03:14

Fichier

vk1.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00938354, version 1

Citation

Ali Khanagha, Vahid Khanagha. Selective Tap Training of FIR filters for Blind Source Separation of Convolutive Speech Mixtures. 2009 IEEE Symposium on Industrial Electronics and Applications (ISIEA 2009), Oct 2009, Kuala Lumpur, Malaysia. 2009. 〈hal-00938354〉

Partager

Métriques

Consultations de la notice

90

Téléchargements de fichiers

60