CLASS-COMBINED ORD N-GRAM FOR ROBUST LANGUAGE ODELING

Norihiko Kobayashi; Tetsunori Kobayashi

CLASS-COMBINED ORD N-GRAM FOR ROBUST LANGUAGE ODELING

Norihiko Kobayashi, Tetsunori Kobayashi

School of Fundamental Science and Engineering

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

1 Citation (Scopus)

Abstract

We propose a method of robust language model ing for a small amount of training text corpus. In this method, the word bigram and the class bigram are combined using a weighting function of preceding word frequency. We made experiments on speech recogni tion using JNAS speech corpus. As the results, it was proved that the performance of the class combined bi gram is equivalent to that of the word bigram trained with 2.5 larger size of corpus. We also made experi ments using sports news dialogue on TV. Recognition accuracy of the class-combined bigram was 83.3% that was 5.5 point higher than that of the word bigram.

Original language	English
Pages	1599-1602
Number of pages	4
Publication status	Published - 1999
Event	6th European Conference on Speech Communication and Technology, EUROSPEECH 1999 - Budapest, Hungary Duration: 1999 Sept 5 → 1999 Sept 9

Conference

Conference	6th European Conference on Speech Communication and Technology, EUROSPEECH 1999
Country/Territory	Hungary
City	Budapest
Period	99/9/5 → 99/9/9

ASJC Scopus subject areas

Computer Science Applications
Software
Linguistics and Language
Communication

Cite this

@conference{e049b9f6adaa4369a314f64a0a0b07f3,

title = "CLASS-COMBINED ORD N-GRAM FOR ROBUST LANGUAGE ODELING",

abstract = "We propose a method of robust language model ing for a small amount of training text corpus. In this method, the word bigram and the class bigram are combined using a weighting function of preceding word frequency. We made experiments on speech recogni tion using JNAS speech corpus. As the results, it was proved that the performance of the class combined bi gram is equivalent to that of the word bigram trained with 2.5 larger size of corpus. We also made experi ments using sports news dialogue on TV. Recognition accuracy of the class-combined bigram was 83.3% that was 5.5 point higher than that of the word bigram.",

author = "Norihiko Kobayashi and Tetsunori Kobayashi",

note = "Publisher Copyright: {\textcopyright} 1999 6th European Conference on Speech Communication and Technology, EUROSPEECH 1999. All rights reserved.; 6th European Conference on Speech Communication and Technology, EUROSPEECH 1999 ; Conference date: 05-09-1999 Through 09-09-1999",

year = "1999",

language = "English",

pages = "1599--1602",

}

TY - CONF

T1 - CLASS-COMBINED ORD N-GRAM FOR ROBUST LANGUAGE ODELING

AU - Kobayashi, Norihiko

AU - Kobayashi, Tetsunori

PY - 1999

Y1 - 1999

N2 - We propose a method of robust language model ing for a small amount of training text corpus. In this method, the word bigram and the class bigram are combined using a weighting function of preceding word frequency. We made experiments on speech recogni tion using JNAS speech corpus. As the results, it was proved that the performance of the class combined bi gram is equivalent to that of the word bigram trained with 2.5 larger size of corpus. We also made experi ments using sports news dialogue on TV. Recognition accuracy of the class-combined bigram was 83.3% that was 5.5 point higher than that of the word bigram.

AB - We propose a method of robust language model ing for a small amount of training text corpus. In this method, the word bigram and the class bigram are combined using a weighting function of preceding word frequency. We made experiments on speech recogni tion using JNAS speech corpus. As the results, it was proved that the performance of the class combined bi gram is equivalent to that of the word bigram trained with 2.5 larger size of corpus. We also made experi ments using sports news dialogue on TV. Recognition accuracy of the class-combined bigram was 83.3% that was 5.5 point higher than that of the word bigram.

UR - http://www.scopus.com/inward/record.url?scp=18644377095&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=18644377095&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:18644377095

SP - 1599

EP - 1602

T2 - 6th European Conference on Speech Communication and Technology, EUROSPEECH 1999

Y2 - 5 September 1999 through 9 September 1999

ER -

CLASS-COMBINED ORD N-GRAM FOR ROBUST LANGUAGE ODELING

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this