NAISTAR
Advanced Search
Japanese | English

naistar (NAIST Academic Repository) >
学術リポジトリ naistar / NAIST Academic Repository naistar >
学術雑誌論文 / Journal Article >
情報科学研究科 / Graduate School of Information Science >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10061/7809

Title: Effect of Central Limit Theorem non-compliance on blind separation of speech by negentropy maximization
Authors: Rajkishore Prasad
Hiroshi Saruwatari
Kiyohiro Shikano
Keywords: Blind signal separation
Independent component analysis
Negentropy
Central limit theorem
Speech signal
Microphone array
Issue Date: Nov-2005
Publisher: 日本音響学会
Journal Title: Acoustical Science and Technology
Volume: 26
Issue: 6
Start page: 511
End page: 522
Abstract: In this paper the blind separation of speech signals from their convoluted mixtures using frequency domain fixed-point independent component analysis algorithm, based on negentropy maximization, is presented. We also discuss fundamental problems of fixed-point ICA by negentropy maximization arising in the separation of the speech signal due to disobedience of the Central Limit Theorem (CLT) by the mixed speech data in the frequency domain. The experimental evidences show that CLT failure is happening due to the spectral sparseness of sources. We also present a blind method to mitigate the negative effects of this by combining null beamforming with the ICA. This combination gives a good result under the low reverberation conditions.
URI: http://hdl.handle.net/10061/7809
ISSN: 1346-3969
Rights: Copyright (C) 2005 日本音響学会.
Text Version: publisher
Publisher DOI: 10.1250/ast.26.511
Appears in Collections:情報科学研究科 / Graduate School of Information Science

Files in This Item:

File SizeFormat
AcouSciTech_26_6_511.pdf6.49 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Copyright (c) 2007-2012 Nara Institute of Science and Technology All Rights Reserved.
DSpace Software Copyright © 2002-2010  Duraspace - Feedback