Performance improvement in single-channel blind separation with pseudo-stereo mixture using cochleagram representation based masks

Yadi Yuan, N. Tengtrairat, W. L. Woo, S. S. Dlay

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A novel single-channel blind source separation (SCBSS) algorithm using Cochleagram-mask based technique is presented in this paper. The proposed system offers benefits such as resemblance of a stereo signal concept given by one microphone, independent of initialization and a priori knowledge of the sources, improved performance with sources which do not strictly satisfy the windowed-disjoint orthogonality (WDO) condition and reduced computational complexity without the need for iterative optimizations. The separation process comprises three steps: 1) estimation of source characteristics, where the source signals are modelled by the autoregressive process and 2) construction of masks using only the single-channel mixture 3) binary TF masks modification through cochleagram processing 4) improve separation depth through independent component analysis (ICA). Experiment results revealed at least 3dB signal to interference ratio comparing to previous systems [1].

Original languageEnglish
Title of host publication2nd IET International Conference on Intelligent Signal Processing 2015 (ISP)
PublisherIEEE
ISBN (Electronic)978-1-78561-137-7
ISBN (Print)978-1-78561-136-0
DOIs
Publication statusPublished - 17 Nov 2016
Event2nd IET International Conference on Intelligent Signal Processing 2015, ISP 2015 - London, United Kingdom
Duration: 1 Dec 20152 Dec 2015

Conference

Conference2nd IET International Conference on Intelligent Signal Processing 2015, ISP 2015
CountryUnited Kingdom
CityLondon
Period1/12/152/12/15

Fingerprint Dive into the research topics of 'Performance improvement in single-channel blind separation with pseudo-stereo mixture using cochleagram representation based masks'. Together they form a unique fingerprint.

Cite this