Paper Info
Title
Improved Wideband Blind Adaptive System Identification Using Decorrelation Filters for the Localization of Multiple Speakers
Abstract
This paper addresses the TDOA extraction problem for localizing multiple sources in noisy and reverberant environments with emphasis on speech excitation. TDOAs are estimated by performing blind adaptive MIMO system identification using a gradient-based BSS variant of the TRINICON framework. We present a novel method to improve the TDOA estimation for signals with lowpass-like spectral characteristics such as speech, for which the standard approach achieves only imperfect system identification. To this end, we propose to combine the BSS algorithm with decorrelation filters, thereby achieving a greatly improved wideband identification of the acoustical system. The approach is verified in a number of scenarios, where it provides more accurate TDOA estimates for the speaker localization at a negligible additional computational cost.
Year
DOI
Venue
2007
10.1109/ISCAS.2007.377971
New Orleans, LA
Keywords
Field
DocType
MIMO systems,acoustic radiators,blind source separation,decorrelation,direction-of-arrival estimation,speaker recognition,time-of-arrival estimation,TDOA extraction problem,TRINICON framework,blind adaptive MIMO system identification,decorrelation filters,gradient-based BSS variant,lowpass-like spectral characteristics,multiple source localization,multiple speaker localization,noisy environment,reverberant environment,speech excitation,time difference of arrival estimation,wideband blind adaptive system identification
Signal processing,Wideband,Decorrelation,Computer science,MIMO,Speech recognition,Electronic engineering,Speaker recognition,Adaptive filter,System identification,Blind signal separation
Conference
ISSN
ISBN
Citations 
0271-4302
1-4244-0921-7
1
PageRank 
References 
Authors
0.38
5
3
Name
Order
Citations
PageRank
Anthony Lombard1517.68
Herbert Buchner243540.57
W. Kellermann368671.03