Staff View: Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech

Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech

This paper presents the endpoint detection approaches specifically for an isolated word uses Malay spoken speeches from Malaysian Parliamentary session. Currently, there are 7,995 vocabularies of utterances in the database collection and for the purpose of this study; the vocabulary is limited to te...

Full description

Saved in:

Bibliographic Details
Main Authors:	Seman N., Bakar Z.A., Bakar N.A., Mohamed H.F., Abdullah N.A.S., Ramakrisnan P., Ahmad S.M.S.
Other Authors:	24825478400
Format:	Conference paper
Published:	2023
Subjects:	Endpoint detection Infinite impulse response Mel frequency cepstral coefficient Short-time energy Short-time zero crossing Frequency response Hidden Markov models Impulse response Information retrieval Knowledge management End point detection Mel-frequency cepstral coefficients Zero-crossings Speech recognition
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my.uniten.dspace-29658
record_format	dspace
spelling	my.uniten.dspace-296582023-12-28T15:17:55Z Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech Seman N. Bakar Z.A. Bakar N.A. Mohamed H.F. Abdullah N.A.S. Ramakrisnan P. Ahmad S.M.S. 24825478400 6507862938 25824639200 57225099928 55433263000 24825389600 24721182400 Endpoint detection Infinite impulse response Mel frequency cepstral coefficient Short-time energy Short-time zero crossing Frequency response Hidden Markov models Impulse response Information retrieval Knowledge management End point detection Infinite impulse response Mel-frequency cepstral coefficients Short-time energy Zero-crossings Speech recognition This paper presents the endpoint detection approaches specifically for an isolated word uses Malay spoken speeches from Malaysian Parliamentary session. Currently, there are 7,995 vocabularies of utterances in the database collection and for the purpose of this study; the vocabulary is limited to ten words which are most frequently spoken selected from ten speakers. Endpoint detection, which aims to distinguish the speech and non-speech segments of digital speech signal, is considered as one of the key preprocessing steps in speech recognition system. Proper estimation of the start and end of the speech (versus silence or background noise) avoids the waste of speech recognition evaluations on preceding or ensuing silence. In this study, the endpoint detection and speech segmentation task is achieved by using the short-time energy (STE) and short-time zero crossing (STZC) measures and combination of both approaches. As a result, the Hidden Markov Model (HMM) recognizer derived the recognition accuracy rate of 91.4% for combination of both algorithms, if compared only 86.3% for STE and 82.1% for STZC rate alone. The experiments show that there are many problems arise where there are still misdetection of word boundaries for the words with weak fricative and nasal sounds. Other obstacles issues such as speaking styles or mood of speaking can also cause the recognition performance. �2010 IEEE. Final 2023-12-28T07:17:55Z 2023-12-28T07:17:55Z 2010 Conference paper 10.1109/INFRKM.2010.5466898 2-s2.0-77953879739 https://www.scopus.com/inward/record.uri?eid=2-s2.0-77953879739&doi=10.1109%2fINFRKM.2010.5466898&partnerID=40&md5=09742bf78ae947336c6796cdf8114a04 https://irepository.uniten.edu.my/handle/123456789/29658 5466898 291 296 Scopus
institution	Universiti Tenaga Nasional
building	UNITEN Library
collection	Institutional Repository
continent	Asia
country	Malaysia
content_provider	Universiti Tenaga Nasional
content_source	UNITEN Institutional Repository
url_provider	http://dspace.uniten.edu.my/
topic	Endpoint detection Infinite impulse response Mel frequency cepstral coefficient Short-time energy Short-time zero crossing Frequency response Hidden Markov models Impulse response Information retrieval Knowledge management End point detection Infinite impulse response Mel-frequency cepstral coefficients Short-time energy Zero-crossings Speech recognition
spellingShingle	Endpoint detection Infinite impulse response Mel frequency cepstral coefficient Short-time energy Short-time zero crossing Frequency response Hidden Markov models Impulse response Information retrieval Knowledge management End point detection Infinite impulse response Mel-frequency cepstral coefficients Short-time energy Zero-crossings Speech recognition Seman N. Bakar Z.A. Bakar N.A. Mohamed H.F. Abdullah N.A.S. Ramakrisnan P. Ahmad S.M.S. Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech
description	This paper presents the endpoint detection approaches specifically for an isolated word uses Malay spoken speeches from Malaysian Parliamentary session. Currently, there are 7,995 vocabularies of utterances in the database collection and for the purpose of this study; the vocabulary is limited to ten words which are most frequently spoken selected from ten speakers. Endpoint detection, which aims to distinguish the speech and non-speech segments of digital speech signal, is considered as one of the key preprocessing steps in speech recognition system. Proper estimation of the start and end of the speech (versus silence or background noise) avoids the waste of speech recognition evaluations on preceding or ensuing silence. In this study, the endpoint detection and speech segmentation task is achieved by using the short-time energy (STE) and short-time zero crossing (STZC) measures and combination of both approaches. As a result, the Hidden Markov Model (HMM) recognizer derived the recognition accuracy rate of 91.4% for combination of both algorithms, if compared only 86.3% for STE and 82.1% for STZC rate alone. The experiments show that there are many problems arise where there are still misdetection of word boundaries for the words with weak fricative and nasal sounds. Other obstacles issues such as speaking styles or mood of speaking can also cause the recognition performance. �2010 IEEE.
author2	24825478400
author_facet	24825478400 Seman N. Bakar Z.A. Bakar N.A. Mohamed H.F. Abdullah N.A.S. Ramakrisnan P. Ahmad S.M.S.
format	Conference paper
author	Seman N. Bakar Z.A. Bakar N.A. Mohamed H.F. Abdullah N.A.S. Ramakrisnan P. Ahmad S.M.S.
author_sort	Seman N.
title	Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech
title_short	Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech
title_full	Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech
title_fullStr	Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech
title_full_unstemmed	Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech
title_sort	evaluating endpoint detection algorithms for isolated word from malay parliamentary speech
publishDate	2023
_version_	1806428181584936960
score	13.222552

Evaluating endpoint detection algorithms for isolated word from Malay parliamentary speech

Similar Items