On the use of contextual time frequency information for full-band clustering-based convolutive blind source separation

In this paper we propose to incorporate contextual time frequency information for clustering-based blind source separation. Previous clustering-based approaches have successfully used clustering techniques to estimate time-frequency separation masks; however, these approaches generally do not consid...

Full description

Bibliographic Details
Main Authors: Atcheson, M., Jafari, I., Togneri, R., Nordholm, Sven
Other Authors: Maria S. Greco, University of Pisa
Format: Conference Paper
Published: IEEE 2014
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/28633
_version_ 1848752588771557376
author Atcheson, M.
Jafari, I.
Togneri, R.
Nordholm, Sven
author2 Maria S. Greco, University of Pisa
author_facet Maria S. Greco, University of Pisa
Atcheson, M.
Jafari, I.
Togneri, R.
Nordholm, Sven
author_sort Atcheson, M.
building Curtin Institutional Repository
collection Online Access
description In this paper we propose to incorporate contextual time frequency information for clustering-based blind source separation. Previous clustering-based approaches have successfully used clustering techniques to estimate time-frequency separation masks; however, these approaches generally do not consider the contextual information of each time-frequency slot. Motivated by the homogenous behavior of speech signals, we modify the fuzzy c-means clustering to bias the results in favor of cluster membership homogeneity within localized neighborhoods in the time-frequency space. Experimental evaluations in both simulated and real-world underdetermined environments demonstrate improvement in source separation performance over previous clustering approaches.
first_indexed 2025-11-14T08:11:01Z
format Conference Paper
id curtin-20.500.11937-28633
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T08:11:01Z
publishDate 2014
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-286332017-09-13T15:17:10Z On the use of contextual time frequency information for full-band clustering-based convolutive blind source separation Atcheson, M. Jafari, I. Togneri, R. Nordholm, Sven Maria S. Greco, University of Pisa fuzzy c-means clustering time-frequency masking blind source separation contextual information In this paper we propose to incorporate contextual time frequency information for clustering-based blind source separation. Previous clustering-based approaches have successfully used clustering techniques to estimate time-frequency separation masks; however, these approaches generally do not consider the contextual information of each time-frequency slot. Motivated by the homogenous behavior of speech signals, we modify the fuzzy c-means clustering to bias the results in favor of cluster membership homogeneity within localized neighborhoods in the time-frequency space. Experimental evaluations in both simulated and real-world underdetermined environments demonstrate improvement in source separation performance over previous clustering approaches. 2014 Conference Paper http://hdl.handle.net/20.500.11937/28633 10.1109/ICASSP.2014.6853972 IEEE restricted
spellingShingle fuzzy c-means clustering
time-frequency masking
blind source separation
contextual information
Atcheson, M.
Jafari, I.
Togneri, R.
Nordholm, Sven
On the use of contextual time frequency information for full-band clustering-based convolutive blind source separation
title On the use of contextual time frequency information for full-band clustering-based convolutive blind source separation
title_full On the use of contextual time frequency information for full-band clustering-based convolutive blind source separation
title_fullStr On the use of contextual time frequency information for full-band clustering-based convolutive blind source separation
title_full_unstemmed On the use of contextual time frequency information for full-band clustering-based convolutive blind source separation
title_short On the use of contextual time frequency information for full-band clustering-based convolutive blind source separation
title_sort on the use of contextual time frequency information for full-band clustering-based convolutive blind source separation
topic fuzzy c-means clustering
time-frequency masking
blind source separation
contextual information
url http://hdl.handle.net/20.500.11937/28633