On the integration of time-frequency masking speech separation and recognition in underdetermined environments

The successful application of automatic speech recognition systems in the real world is conditional on its ability to handle realistic environments with unfavorable conditions such as reverberation and multiple sources of inteference. Previous research has identified time-frequency masking based app...

Full description

Bibliographic Details
Main Authors: Jafari, I., Haque, S., Togneri, R., Nordholm, Sven
Format: Conference Paper
Published: 2012
Online Access:http://hdl.handle.net/20.500.11937/5370
_version_ 1848744778080976896
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 The successful application of automatic speech recognition systems in the real world is conditional on its ability to handle realistic environments with unfavorable conditions such as reverberation and multiple sources of inteference. Previous research has identified time-frequency masking based approaches to blind source separation as a viable approach for multisource reverberant source separation. It is proposed the use of such separation techniques as a front-end to speech recognition will encourage greater recognition accuracy. Experimental evaluations confirmed the hypothesis with an improvement in recognition accuracy of over 20% at a reverberation time of RT60 = 300ms; this is indicative of the potential for future research in this field. © 2012 IEEE.
first_indexed 2025-11-14T06:06:52Z
format Conference Paper
id curtin-20.500.11937-5370
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T06:06:52Z
publishDate 2012
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-53702017-09-13T14:39:54Z On the integration of time-frequency masking speech separation and recognition in underdetermined environments Jafari, I. Haque, S. Togneri, R. Nordholm, Sven The successful application of automatic speech recognition systems in the real world is conditional on its ability to handle realistic environments with unfavorable conditions such as reverberation and multiple sources of inteference. Previous research has identified time-frequency masking based approaches to blind source separation as a viable approach for multisource reverberant source separation. It is proposed the use of such separation techniques as a front-end to speech recognition will encourage greater recognition accuracy. Experimental evaluations confirmed the hypothesis with an improvement in recognition accuracy of over 20% at a reverberation time of RT60 = 300ms; this is indicative of the potential for future research in this field. © 2012 IEEE. 2012 Conference Paper http://hdl.handle.net/20.500.11937/5370 10.1109/ACSSC.2012.6489303 restricted
spellingShingle Jafari, I.
Haque, S.
Togneri, R.
Nordholm, Sven
On the integration of time-frequency masking speech separation and recognition in underdetermined environments
title On the integration of time-frequency masking speech separation and recognition in underdetermined environments
title_full On the integration of time-frequency masking speech separation and recognition in underdetermined environments
title_fullStr On the integration of time-frequency masking speech separation and recognition in underdetermined environments
title_full_unstemmed On the integration of time-frequency masking speech separation and recognition in underdetermined environments
title_short On the integration of time-frequency masking speech separation and recognition in underdetermined environments
title_sort on the integration of time-frequency masking speech separation and recognition in underdetermined environments
url http://hdl.handle.net/20.500.11937/5370