Multichannel blind deconvolution for source separation in convolutive mixtures of speech

Authors:
K. Kokkinakis;A. K. Nandi
Affiliations:
Signal Process. & Commun. Group, Univ. of Liverpool, UK;-
Venue:
IEEE Transactions on Audio, Speech, and Language Processing
Year:
2006

Citing 0
Cited 6

Letters: Binaural semi-blind dereverberation of noisy convoluted speech signals

Neurocomputing
Signal separation by integrating adaptive beamforming with blind deconvolution

ICA'07 Proceedings of the 7th international conference on Independent component analysis and signal separation
Multichannel blind separation and deconvolution of images for document analysis

IEEE Transactions on Image Processing
Blind source separation with low frequency compensation for convolutive mixtures

Asilomar'09 Proceedings of the 43rd Asilomar conference on Signals, systems and computers
Correlation-based amplitude estimation of coincident partials in monaural musical signals

EURASIP Journal on Audio, Speech, and Music Processing
Efficient blind dereverberation and echo cancellation based on independent component analysis for actual acoustic signals

Neural Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper addresses the blind separation of convolutive and temporally correlated mixtures of speech, through the use of a multichannel blind deconvolution (MBD) method. In the proposed framework (LP-NGA), spatio-temporal separation is carried out by entropy maximization using the well-known natural gradient algorithm (NGA), while a temporal pre-whitening stage, based on linear prediction (LP), manages to fully preserve the original spectral characteristics of each source contribution. Confronted with synthetic convolutive mixtures, we show that the LP-NGA-an unconstrained natural extension to the multichannel BSS problem-benefits not only from fewer model constraints, but also from other factors, such as an overall increase in separation performance, spectral preservation efficiency and speed of convergence.