An embedded knowledge integration for hybrid language modeling

Shuwu Zhang, Hirofumi Yamamoto, Yoshinori Sagisaka

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper describes an embedded architecture to couple utilizable language knowledge and innovative language models, as well as modeling approaches, for intensive language modeling in speech recognition. In this embedded mechanism, three innovative language modeling approaches at different levels, ie., composite N-gram, dis tance-related unit association maximum entropy (DU-AME), and linkgram, have different functions to extend the definitions of basic language units, favorably improve the underlying model instead of conventional N-grams and provide effective combination with longer history syntactic lnk dependency knowledge, respectively. In this threelevel hybrid language modeling, each lower level modeling serves the higher level modeling(s). The results in each level are well utized or embedded in the higher level(s). These models can be trained level by level Accordingly, some prospective language constraints can finally be embedded in a wellorganized hybrid model. Experimental data based on the embedded modeling show that the hybrid model reduces WER 14.5% compared with the conventional word-based bigram model As a result, it can be expected to improve the conventional statistical language modelng.

Original languageEnglish
Title of host publication6th International Conference on Spoken Language Processing, ICSLP 2000
PublisherInternational Speech Communication Association
ISBN (Electronic)7801501144, 9787801501141
Publication statusPublished - 2000
Externally publishedYes
Event6th International Conference on Spoken Language Processing, ICSLP 2000 - Beijing, China
Duration: 2000 Oct 162000 Oct 20

Other

Other6th International Conference on Spoken Language Processing, ICSLP 2000
CountryChina
CityBeijing
Period00/10/1600/10/20

Fingerprint

language
knowledge
Language Modeling
Language
entropy
Modeling
Conventional
N-gram
Hybrid Model
history
History
Maximum Entropy
Speech Recognition
Syntactic Dependency
Language Model

ASJC Scopus subject areas

  • Linguistics and Language
  • Language and Linguistics

Cite this

Zhang, S., Yamamoto, H., & Sagisaka, Y. (2000). An embedded knowledge integration for hybrid language modeling. In 6th International Conference on Spoken Language Processing, ICSLP 2000 International Speech Communication Association.

An embedded knowledge integration for hybrid language modeling. / Zhang, Shuwu; Yamamoto, Hirofumi; Sagisaka, Yoshinori.

6th International Conference on Spoken Language Processing, ICSLP 2000. International Speech Communication Association, 2000.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhang, S, Yamamoto, H & Sagisaka, Y 2000, An embedded knowledge integration for hybrid language modeling. in 6th International Conference on Spoken Language Processing, ICSLP 2000. International Speech Communication Association, 6th International Conference on Spoken Language Processing, ICSLP 2000, Beijing, China, 00/10/16.
Zhang S, Yamamoto H, Sagisaka Y. An embedded knowledge integration for hybrid language modeling. In 6th International Conference on Spoken Language Processing, ICSLP 2000. International Speech Communication Association. 2000
Zhang, Shuwu ; Yamamoto, Hirofumi ; Sagisaka, Yoshinori. / An embedded knowledge integration for hybrid language modeling. 6th International Conference on Spoken Language Processing, ICSLP 2000. International Speech Communication Association, 2000.
@inproceedings{9d101bde06e74b2e89652f89611ce08d,
title = "An embedded knowledge integration for hybrid language modeling",
abstract = "This paper describes an embedded architecture to couple utilizable language knowledge and innovative language models, as well as modeling approaches, for intensive language modeling in speech recognition. In this embedded mechanism, three innovative language modeling approaches at different levels, ie., composite N-gram, dis tance-related unit association maximum entropy (DU-AME), and linkgram, have different functions to extend the definitions of basic language units, favorably improve the underlying model instead of conventional N-grams and provide effective combination with longer history syntactic lnk dependency knowledge, respectively. In this threelevel hybrid language modeling, each lower level modeling serves the higher level modeling(s). The results in each level are well utized or embedded in the higher level(s). These models can be trained level by level Accordingly, some prospective language constraints can finally be embedded in a wellorganized hybrid model. Experimental data based on the embedded modeling show that the hybrid model reduces WER 14.5{\%} compared with the conventional word-based bigram model As a result, it can be expected to improve the conventional statistical language modelng.",
author = "Shuwu Zhang and Hirofumi Yamamoto and Yoshinori Sagisaka",
year = "2000",
language = "English",
booktitle = "6th International Conference on Spoken Language Processing, ICSLP 2000",
publisher = "International Speech Communication Association",

}

TY - GEN

T1 - An embedded knowledge integration for hybrid language modeling

AU - Zhang, Shuwu

AU - Yamamoto, Hirofumi

AU - Sagisaka, Yoshinori

PY - 2000

Y1 - 2000

N2 - This paper describes an embedded architecture to couple utilizable language knowledge and innovative language models, as well as modeling approaches, for intensive language modeling in speech recognition. In this embedded mechanism, three innovative language modeling approaches at different levels, ie., composite N-gram, dis tance-related unit association maximum entropy (DU-AME), and linkgram, have different functions to extend the definitions of basic language units, favorably improve the underlying model instead of conventional N-grams and provide effective combination with longer history syntactic lnk dependency knowledge, respectively. In this threelevel hybrid language modeling, each lower level modeling serves the higher level modeling(s). The results in each level are well utized or embedded in the higher level(s). These models can be trained level by level Accordingly, some prospective language constraints can finally be embedded in a wellorganized hybrid model. Experimental data based on the embedded modeling show that the hybrid model reduces WER 14.5% compared with the conventional word-based bigram model As a result, it can be expected to improve the conventional statistical language modelng.

AB - This paper describes an embedded architecture to couple utilizable language knowledge and innovative language models, as well as modeling approaches, for intensive language modeling in speech recognition. In this embedded mechanism, three innovative language modeling approaches at different levels, ie., composite N-gram, dis tance-related unit association maximum entropy (DU-AME), and linkgram, have different functions to extend the definitions of basic language units, favorably improve the underlying model instead of conventional N-grams and provide effective combination with longer history syntactic lnk dependency knowledge, respectively. In this threelevel hybrid language modeling, each lower level modeling serves the higher level modeling(s). The results in each level are well utized or embedded in the higher level(s). These models can be trained level by level Accordingly, some prospective language constraints can finally be embedded in a wellorganized hybrid model. Experimental data based on the embedded modeling show that the hybrid model reduces WER 14.5% compared with the conventional word-based bigram model As a result, it can be expected to improve the conventional statistical language modelng.

UR - http://www.scopus.com/inward/record.url?scp=85009080610&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009080610&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85009080610

BT - 6th International Conference on Spoken Language Processing, ICSLP 2000

PB - International Speech Communication Association

ER -