Parsing continuous speech by HMM-LR method

Kenji Kita; Takeshi Kawabata; Hiroaki Saito

Parsing continuous speech by HMM-LR method

Kenji Kita, Takeshi Kawabata, Hiroaki Saito

Research output: Contribution to conference › Paper › peer-review

Abstract

This paper describes a speech parsing method called HMM-LR. In HMM-LR, an LR parsing table is used to predict phones in speech input, and the system drives an HMM-based speech recognizer directly without any intervening structures such as a phone lattice. Very accurate, efficient speech parsing is achieved through the integrated processes of speech recognition and language analysis. The HMM-LR method is applied to large-vocabulary speaker-dependent Japanese phrase recognition. The recognition rate is 87.1% for the top candidates and 97.7% for the five best candidates.

Original language	English
Pages	126-131
Number of pages	6
Publication status	Published - 1989
Externally published	Yes
Event	1st International Workshop on Parsing Technologies, IWPT 1989 - Pittsburgh, United States Duration: 1989 Aug 28 → 1989 Aug 31

Conference

Conference	1st International Workshop on Parsing Technologies, IWPT 1989
Country/Territory	United States
City	Pittsburgh
Period	89/8/28 → 89/8/31

ASJC Scopus subject areas

Linguistics and Language
Language and Linguistics
Computer Science Applications

Cite this

@conference{82c207ebf1f2447eb6ca98cb3fa26785,

title = "Parsing continuous speech by HMM-LR method",

abstract = "This paper describes a speech parsing method called HMM-LR. In HMM-LR, an LR parsing table is used to predict phones in speech input, and the system drives an HMM-based speech recognizer directly without any intervening structures such as a phone lattice. Very accurate, efficient speech parsing is achieved through the integrated processes of speech recognition and language analysis. The HMM-LR method is applied to large-vocabulary speaker-dependent Japanese phrase recognition. The recognition rate is 87.1% for the top candidates and 97.7% for the five best candidates.",

author = "Kenji Kita and Takeshi Kawabata and Hiroaki Saito",

note = "Publisher Copyright: Copyright {\textcopyright} 1989 Massachusetts Institute of Technology All Rights Reserved; 1st International Workshop on Parsing Technologies, IWPT 1989 ; Conference date: 28-08-1989 Through 31-08-1989",

year = "1989",

language = "English",

pages = "126--131",

}

TY - CONF

T1 - Parsing continuous speech by HMM-LR method

AU - Kita, Kenji

AU - Kawabata, Takeshi

AU - Saito, Hiroaki

PY - 1989

Y1 - 1989

N2 - This paper describes a speech parsing method called HMM-LR. In HMM-LR, an LR parsing table is used to predict phones in speech input, and the system drives an HMM-based speech recognizer directly without any intervening structures such as a phone lattice. Very accurate, efficient speech parsing is achieved through the integrated processes of speech recognition and language analysis. The HMM-LR method is applied to large-vocabulary speaker-dependent Japanese phrase recognition. The recognition rate is 87.1% for the top candidates and 97.7% for the five best candidates.

AB - This paper describes a speech parsing method called HMM-LR. In HMM-LR, an LR parsing table is used to predict phones in speech input, and the system drives an HMM-based speech recognizer directly without any intervening structures such as a phone lattice. Very accurate, efficient speech parsing is achieved through the integrated processes of speech recognition and language analysis. The HMM-LR method is applied to large-vocabulary speaker-dependent Japanese phrase recognition. The recognition rate is 87.1% for the top candidates and 97.7% for the five best candidates.

UR - http://www.scopus.com/inward/record.url?scp=85119656990&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85119656990&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85119656990

SP - 126

EP - 131

T2 - 1st International Workshop on Parsing Technologies, IWPT 1989

Y2 - 28 August 1989 through 31 August 1989

ER -

Parsing continuous speech by HMM-LR method

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this