A novel fuzzy clustering algorithm using observation weighting and context information for reverberant blind speech separation

Time-frequency masking has evolved as a powerful tool for tackling blind source separation problems. In previous work, mask estimation was performed with the help of well-known standard cluster algorithms. Spatial observation vectors, extracted from a set of microphones, were grouped into separate c...

Full description

Bibliographic Details
Main Authors: Kuhne, M., Togneri, R., Nordholm, Sven
Format: Journal Article
Published: Elsevier Science 2009
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/44787
_version_ 1848757101664403456
author Kuhne, M.
Togneri, R.
Nordholm, Sven
author_facet Kuhne, M.
Togneri, R.
Nordholm, Sven
author_sort Kuhne, M.
building Curtin Institutional Repository
collection Online Access
description Time-frequency masking has evolved as a powerful tool for tackling blind source separation problems. In previous work, mask estimation was performed with the help of well-known standard cluster algorithms. Spatial observation vectors, extracted from a set of microphones, were grouped into separate clusters, each representing a particular source. However, most off-the-shelf clustering methods are not very robust to outliers or noise in the data.This lack of robustness often leads to in correct localization and partitioning results, particularly for reverberant speech mixtures.To address this issue, we investigate the use of observation weights and context information as means to improve the clustering performance under reverberant conditions. While the observation weights improve the localization accuracy by ignoring noisy observations, context information smoothes the cluster membership levels by exploiting the highly structured nature of speech signals in the time-frequency domain. In a number of experiments, we demonstrate the superiority of the proposed method over conventional fuzzy clustering, both in terms of localization accuracy as well as speech separation performance.
first_indexed 2025-11-14T09:22:45Z
format Journal Article
id curtin-20.500.11937-44787
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T09:22:45Z
publishDate 2009
publisher Elsevier Science
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-447872017-09-13T16:01:01Z A novel fuzzy clustering algorithm using observation weighting and context information for reverberant blind speech separation Kuhne, M. Togneri, R. Nordholm, Sven Blind source separation Time-frequency masking Fuzzy clustering Reverberation Robustness Adaptive beamforming Time-frequency masking has evolved as a powerful tool for tackling blind source separation problems. In previous work, mask estimation was performed with the help of well-known standard cluster algorithms. Spatial observation vectors, extracted from a set of microphones, were grouped into separate clusters, each representing a particular source. However, most off-the-shelf clustering methods are not very robust to outliers or noise in the data.This lack of robustness often leads to in correct localization and partitioning results, particularly for reverberant speech mixtures.To address this issue, we investigate the use of observation weights and context information as means to improve the clustering performance under reverberant conditions. While the observation weights improve the localization accuracy by ignoring noisy observations, context information smoothes the cluster membership levels by exploiting the highly structured nature of speech signals in the time-frequency domain. In a number of experiments, we demonstrate the superiority of the proposed method over conventional fuzzy clustering, both in terms of localization accuracy as well as speech separation performance. 2009 Journal Article http://hdl.handle.net/20.500.11937/44787 10.1016/j.sigpro.2009.08.005 Elsevier Science fulltext
spellingShingle Blind source separation
Time-frequency masking
Fuzzy clustering
Reverberation Robustness
Adaptive beamforming
Kuhne, M.
Togneri, R.
Nordholm, Sven
A novel fuzzy clustering algorithm using observation weighting and context information for reverberant blind speech separation
title A novel fuzzy clustering algorithm using observation weighting and context information for reverberant blind speech separation
title_full A novel fuzzy clustering algorithm using observation weighting and context information for reverberant blind speech separation
title_fullStr A novel fuzzy clustering algorithm using observation weighting and context information for reverberant blind speech separation
title_full_unstemmed A novel fuzzy clustering algorithm using observation weighting and context information for reverberant blind speech separation
title_short A novel fuzzy clustering algorithm using observation weighting and context information for reverberant blind speech separation
title_sort novel fuzzy clustering algorithm using observation weighting and context information for reverberant blind speech separation
topic Blind source separation
Time-frequency masking
Fuzzy clustering
Reverberation Robustness
Adaptive beamforming
url http://hdl.handle.net/20.500.11937/44787