Enhanced Running Spectrum Analysis for Robust Speech Recognition Under Adverse Conditions: A Case Study on Japanese Speech

Mufungulwa, G; Tsutsui, H; Miyanaga, Y; Abe, S-I

Enhanced Running Spectrum Analysis for Robust Speech Recognition Under Adverse Conditions: A Case Study on Japanese Speech

Mufungulwa, G Tsutsui, H Miyanaga, Y

Abe, S-I

Permalink

Publisher:: ECTI
Publication Type:: Journal Article
Citation:: ECTI Transactions on Computer and Information Technology (ECTI-CIT), 2017, 11, (1), pp. 82-90
Issue Date:: 2017-07-02

Closed Access

	Filename	Description	Size
	document (1).pdf	Published version	667.05 kB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Mufungulwa, G
dc.contributor.author	Tsutsui, H
dc.contributor.author	Miyanaga, Y https://orcid.org/0000-0002-2795-2234
dc.contributor.author	Abe, S-I
dc.date.accessioned	2022-09-22T20:20:19Z
dc.date.available	2022-09-22T20:20:19Z
dc.date.issued	2017-07-02
dc.identifier.citation	ECTI Transactions on Computer and Information Technology (ECTI-CIT), 2017, 11, (1), pp. 82-90
dc.identifier.issn	2286-9131
dc.identifier.issn	2286-9131
dc.identifier.uri	http://hdl.handle.net/10453/161999
dc.description.abstract	<jats:p>In any real environment, noises degrade the performance of Automatic Speech Recognition (ASR) systems. Additionally, in the case of similar pronunciations, it is not easy to realize a high accuracy of recognition. From this point of view, our work envisions an enhanced algorithm processing a speech modulation spectrum, such as Running Spectrum Analysis (RSA). It was also adequately applied to observed speech data. In the envisioned method, a modulation spectrum filtering (MSF) method directly modified the observed cepstral modulation spectrum by a Fourier transform of the cepstral time frequency. The method and experiments carried out for various passbands had favorable results that showed an improvement of about 1-4 % in recognition accuracycompared to conventional methods.</jats:p>
dc.language	en
dc.publisher	ECTI
dc.relation.ispartof	ECTI Transactions on Computer and Information Technology (ECTI-CIT)
dc.relation.isbasedon	10.37936/ecti-cit.2017111.81945
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	Enhanced Running Spectrum Analysis for Robust Speech Recognition Under Adverse Conditions: A Case Study on Japanese Speech
dc.type	Journal Article
utslib.citation.volume	11
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Electrical and Data Engineering
utslib.copyright.status	closed_access	*
dc.date.updated	2022-09-22T20:20:18Z
pubs.issue	1
pubs.publication-status	Published online
pubs.volume	11
utslib.citation.issue	1

Abstract:

In any real environment, noises degrade the performance of Automatic Speech Recognition (ASR) systems. Additionally, in the case of similar pronunciations, it is not easy to realize a high accuracy of recognition. From this point of view, our work envisions an enhanced algorithm processing a speech modulation spectrum, such as Running Spectrum Analysis (RSA). It was also adequately applied to observed speech data. In the envisioned method, a modulation spectrum filtering (MSF) method directly modified the observed cepstral modulation spectrum by a Fourier transform of the cepstral time frequency. The method and experiments carried out for various passbands had favorable results that showed an improvement of about 1-4 % in recognition accuracycompared to conventional methods.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/161999