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:

Title: Subband-Based Blind Separation for Convolutive Mixtures of Speech
Authors: Shoko Araki
Shoji Makino
Robert Aichner
Tsuyoki Nishikawa
Hiroshi Saruwatari
Keywords: blind source separation
speech separation
convolutative mixtures
subband processing
frequency dependent processing
Issue Date: Dec-2005
Publisher: 電子情報通信学会
Journal Title: IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
Volume: E88-A
Issue: 12
Start page: 3593
End page: 3603
Abstract: We propose utilizing subband-based blind source separation (BSS) for convolutive mixtures of speech. This is motivated by the drawback of frequency-domain BSS, i.e., when a long frame with a fixed long frame-shift is used to cover reverberation, the number of samples in each frequency decreases and the separation performance is degraded. In subband BSS, (1) by using a moderate number of subbands, a sufficient number of samples can be held in each subband, and (2) by using FIR filters in each subband, we can manage long reverberation. We confirm that subband BSS achieves better performance than frequency-domain BSS. Moreover, subband BSS allows us to select a separation method suited to each subband. Using this advantage, we propose efficient separation procedures that consider the frequency characteristics of room reverberation and speech signals (3) by using longer unmixing filters in low frequency bands and (4) by adopting an overlap-blockshift in BSS's batch adaptation in low frequency bands. Consequently, frequency-dependent subband processing is successfully realized with the proposed subband BSS.
ISSN: 0916-8508
Rights: Copyright (C) 2005 電子情報通信学会.
Text Version: publisher
Publisher DOI: 10.1093/ietfec/e88-a.12.3593
Appears in Collections:情報科学研究科 / Graduate School of Information Science

Files in This Item:

File SizeFormat
IEICETransFunEleCom_E88A_12_3593.pdf7.92 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