Underdetermined Blind Source Separation with Fuzzy Clustering for Arbitrarily Arranged Sensors

Recently, the concept of time-frequency masking has developed as an important approach to the blind source separation problem, particularly when in the presence of reverberation. However, previous research has been limited by factors such as the sensor arrangement and/or the mask estimation tech...

Full description

Bibliographic Details
Main Authors: Jafari, I., Haque, S., Togneri, R., Nordholm, Sven
Format: Conference Paper
Published: 2011
Online Access:http://hdl.handle.net/20.500.11937/52862
_version_ 1848759029624471552
author Jafari, I.
Haque, S.
Togneri, R.
Nordholm, Sven
author_facet Jafari, I.
Haque, S.
Togneri, R.
Nordholm, Sven
author_sort Jafari, I.
building Curtin Institutional Repository
collection Online Access
description Recently, the concept of time-frequency masking has developed as an important approach to the blind source separation problem, particularly when in the presence of reverberation. However, previous research has been limited by factors such as the sensor arrangement and/or the mask estimation technique implemented. This paper presents a novel integration of two established approaches to BSS in an effort to overcome such limitations. A multidimensional feature vector is extracted from a non-linear sensor arrangement, and the fuzzy c-means algorithm is then applied to cluster the feature vectors into representations of the source speakers. Fuzzy time-frequency masks are estimated and applied to the observations for source recovery. The evaluations on the proposed study demonstrated improved separation quality over all test conditions. This establishes the potential of multidimensional fuzzy c-means clustering for mask estimation in the context of blind source separation
first_indexed 2025-11-14T09:53:23Z
format Conference Paper
id curtin-20.500.11937-52862
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T09:53:23Z
publishDate 2011
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-528622018-08-20T01:00:42Z Underdetermined Blind Source Separation with Fuzzy Clustering for Arbitrarily Arranged Sensors Jafari, I. Haque, S. Togneri, R. Nordholm, Sven Recently, the concept of time-frequency masking has developed as an important approach to the blind source separation problem, particularly when in the presence of reverberation. However, previous research has been limited by factors such as the sensor arrangement and/or the mask estimation technique implemented. This paper presents a novel integration of two established approaches to BSS in an effort to overcome such limitations. A multidimensional feature vector is extracted from a non-linear sensor arrangement, and the fuzzy c-means algorithm is then applied to cluster the feature vectors into representations of the source speakers. Fuzzy time-frequency masks are estimated and applied to the observations for source recovery. The evaluations on the proposed study demonstrated improved separation quality over all test conditions. This establishes the potential of multidimensional fuzzy c-means clustering for mask estimation in the context of blind source separation 2011 Conference Paper http://hdl.handle.net/20.500.11937/52862 fulltext
spellingShingle Jafari, I.
Haque, S.
Togneri, R.
Nordholm, Sven
Underdetermined Blind Source Separation with Fuzzy Clustering for Arbitrarily Arranged Sensors
title Underdetermined Blind Source Separation with Fuzzy Clustering for Arbitrarily Arranged Sensors
title_full Underdetermined Blind Source Separation with Fuzzy Clustering for Arbitrarily Arranged Sensors
title_fullStr Underdetermined Blind Source Separation with Fuzzy Clustering for Arbitrarily Arranged Sensors
title_full_unstemmed Underdetermined Blind Source Separation with Fuzzy Clustering for Arbitrarily Arranged Sensors
title_short Underdetermined Blind Source Separation with Fuzzy Clustering for Arbitrarily Arranged Sensors
title_sort underdetermined blind source separation with fuzzy clustering for arbitrarily arranged sensors
url http://hdl.handle.net/20.500.11937/52862