Subband-based blind separation for convolutive mixtures of speech

Shoko Araki*, Shoji Makino, Robert Aichner, Tsuyoki Nishikawa, Hiroshi Saruwatari

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

21 Citations (Scopus)


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.

Original languageEnglish
Pages (from-to)3593-3602
Number of pages10
JournalIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
Issue number12
Publication statusPublished - 2005 Dec
Externally publishedYes


  • Blind source separation
  • Convolutive mixtures
  • Frequency dependent processing
  • Speech separation
  • Subband processing

ASJC Scopus subject areas

  • Signal Processing
  • Computer Graphics and Computer-Aided Design
  • Electrical and Electronic Engineering
  • Applied Mathematics


Dive into the research topics of 'Subband-based blind separation for convolutive mixtures of speech'. Together they form a unique fingerprint.

Cite this