Three different LR parsing algorithms for phoneme-context-dependent HMM-based continuous speech recognition

Akito Nagai, Shigeki Sagayama, Kenji Kita, Hideaki Kikuchi

Research output: Contribution to journalArticle

Abstract

This paper discusses three approaches for combining an efficient LR parser and phoneme-context-dependent HMMs and compares them through continuous speech recognition experiments. In continuous speech recognition, phoneme-context-dependent allophonic models are considered very helpful for enhancing the recognition accuracy. They precisely represent allophonic variations caused by the difference in phoneme-contexts. With grammatical constraints based on a context free grammar(CFG), a generalized LR parser is one of the most efficient parsing algorithms for speech recognition. Therefore, the combination of allophonic models and a generalized LR parser is a powerful scheme enabling accurate and efficient speech recognition. In this paper, three phoneme-context-dependent LR parsing algorithms are proposed, which make it possible to drive allophonic HMMs. The algorithms are outlined as follows: (1) Algorithm for predicting the phonemic context dynamically in the LR parser using a phoneme-context-independent LR table. (2) Algorithm for converting an LR table into a phoneme-context-dependent LR table. (3) Algorithm for converting a CFG into a phoneme-context-dependent CFG. This paper also includes discussion of the results of recognition experiments, and a comparison of performance and efficiency of these three algorithms.

Original languageEnglish
Pages (from-to)29-37
Number of pages9
JournalIEICE Transactions on Information and Systems
VolumeE76-D
Issue number1
Publication statusPublished - 1993 Jan
Externally publishedYes

Fingerprint

Continuous speech recognition
Context free grammars
Speech recognition
Experiments

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Information Systems
  • Software

Cite this

Three different LR parsing algorithms for phoneme-context-dependent HMM-based continuous speech recognition. / Nagai, Akito; Sagayama, Shigeki; Kita, Kenji; Kikuchi, Hideaki.

In: IEICE Transactions on Information and Systems, Vol. E76-D, No. 1, 01.1993, p. 29-37.

Research output: Contribution to journalArticle

@article{2984d6524f81496e9e676e9bdef6c893,
title = "Three different LR parsing algorithms for phoneme-context-dependent HMM-based continuous speech recognition",
abstract = "This paper discusses three approaches for combining an efficient LR parser and phoneme-context-dependent HMMs and compares them through continuous speech recognition experiments. In continuous speech recognition, phoneme-context-dependent allophonic models are considered very helpful for enhancing the recognition accuracy. They precisely represent allophonic variations caused by the difference in phoneme-contexts. With grammatical constraints based on a context free grammar(CFG), a generalized LR parser is one of the most efficient parsing algorithms for speech recognition. Therefore, the combination of allophonic models and a generalized LR parser is a powerful scheme enabling accurate and efficient speech recognition. In this paper, three phoneme-context-dependent LR parsing algorithms are proposed, which make it possible to drive allophonic HMMs. The algorithms are outlined as follows: (1) Algorithm for predicting the phonemic context dynamically in the LR parser using a phoneme-context-independent LR table. (2) Algorithm for converting an LR table into a phoneme-context-dependent LR table. (3) Algorithm for converting a CFG into a phoneme-context-dependent CFG. This paper also includes discussion of the results of recognition experiments, and a comparison of performance and efficiency of these three algorithms.",
author = "Akito Nagai and Shigeki Sagayama and Kenji Kita and Hideaki Kikuchi",
year = "1993",
month = "1",
language = "English",
volume = "E76-D",
pages = "29--37",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "1",

}

TY - JOUR

T1 - Three different LR parsing algorithms for phoneme-context-dependent HMM-based continuous speech recognition

AU - Nagai, Akito

AU - Sagayama, Shigeki

AU - Kita, Kenji

AU - Kikuchi, Hideaki

PY - 1993/1

Y1 - 1993/1

N2 - This paper discusses three approaches for combining an efficient LR parser and phoneme-context-dependent HMMs and compares them through continuous speech recognition experiments. In continuous speech recognition, phoneme-context-dependent allophonic models are considered very helpful for enhancing the recognition accuracy. They precisely represent allophonic variations caused by the difference in phoneme-contexts. With grammatical constraints based on a context free grammar(CFG), a generalized LR parser is one of the most efficient parsing algorithms for speech recognition. Therefore, the combination of allophonic models and a generalized LR parser is a powerful scheme enabling accurate and efficient speech recognition. In this paper, three phoneme-context-dependent LR parsing algorithms are proposed, which make it possible to drive allophonic HMMs. The algorithms are outlined as follows: (1) Algorithm for predicting the phonemic context dynamically in the LR parser using a phoneme-context-independent LR table. (2) Algorithm for converting an LR table into a phoneme-context-dependent LR table. (3) Algorithm for converting a CFG into a phoneme-context-dependent CFG. This paper also includes discussion of the results of recognition experiments, and a comparison of performance and efficiency of these three algorithms.

AB - This paper discusses three approaches for combining an efficient LR parser and phoneme-context-dependent HMMs and compares them through continuous speech recognition experiments. In continuous speech recognition, phoneme-context-dependent allophonic models are considered very helpful for enhancing the recognition accuracy. They precisely represent allophonic variations caused by the difference in phoneme-contexts. With grammatical constraints based on a context free grammar(CFG), a generalized LR parser is one of the most efficient parsing algorithms for speech recognition. Therefore, the combination of allophonic models and a generalized LR parser is a powerful scheme enabling accurate and efficient speech recognition. In this paper, three phoneme-context-dependent LR parsing algorithms are proposed, which make it possible to drive allophonic HMMs. The algorithms are outlined as follows: (1) Algorithm for predicting the phonemic context dynamically in the LR parser using a phoneme-context-independent LR table. (2) Algorithm for converting an LR table into a phoneme-context-dependent LR table. (3) Algorithm for converting a CFG into a phoneme-context-dependent CFG. This paper also includes discussion of the results of recognition experiments, and a comparison of performance and efficiency of these three algorithms.

UR - http://www.scopus.com/inward/record.url?scp=0027342048&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027342048&partnerID=8YFLogxK

M3 - Article

VL - E76-D

SP - 29

EP - 37

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 1

ER -