A Statistical Language Model for Conventional Speech Reflecting the Previous Utterance of the Other Participant

Hirofumi Yamamoto, Koichi Tanigaki, Yoshinori Sagisaka

Research output: Contribution to journalArticle

Abstract

This paper proposes the construction of a language model for speech recognition in speech translation. The proposed language model is intended to improve speech recognition performance by taking account of the previous utterance content of the counterpart speaker. The previous utterance content of the counterpart speaker is represented by the intermediate language which is generally used by C-star (Spoken Language Translation Research Group). The proposed model is based on a language model that is dependent on the intermediate language representation. An experiment concerning the conversation of hotel reservations shows that the rate of incorrect word recognition is reduced by approximately 5.4% (from 14.7% to 13.9%), and the usefulness of the proposed model is demonstrated.

Original languageEnglish
Pages (from-to)54-62
Number of pages9
JournalElectronics and Communications in Japan, Part II: Electronics (English translation of Denshi Tsushin Gakkai Ronbunshi)
Volume87
Issue number1
DOIs
Publication statusPublished - 2004 Jan

Fingerprint

Speech recognition
speech recognition
Hotels
conversation
carbon stars
Stars
Experiments

Keywords

  • Conversation
  • Intermediate language representation
  • N-gram
  • Statistical language model
  • Task adaptation

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

@article{d3e89cbbd99444bc85abf9419c2761c8,
title = "A Statistical Language Model for Conventional Speech Reflecting the Previous Utterance of the Other Participant",
abstract = "This paper proposes the construction of a language model for speech recognition in speech translation. The proposed language model is intended to improve speech recognition performance by taking account of the previous utterance content of the counterpart speaker. The previous utterance content of the counterpart speaker is represented by the intermediate language which is generally used by C-star (Spoken Language Translation Research Group). The proposed model is based on a language model that is dependent on the intermediate language representation. An experiment concerning the conversation of hotel reservations shows that the rate of incorrect word recognition is reduced by approximately 5.4{\%} (from 14.7{\%} to 13.9{\%}), and the usefulness of the proposed model is demonstrated.",
keywords = "Conversation, Intermediate language representation, N-gram, Statistical language model, Task adaptation",
author = "Hirofumi Yamamoto and Koichi Tanigaki and Yoshinori Sagisaka",
year = "2004",
month = "1",
doi = "10.1002/ecjb.10211",
language = "English",
volume = "87",
pages = "54--62",
journal = "Electronics and Communications in Japan, Part II: Electronics (English translation of Denshi Tsushin Gakkai Ronbunshi)",
issn = "8756-663X",
publisher = "Scripta Technica",
number = "1",

}

TY - JOUR

T1 - A Statistical Language Model for Conventional Speech Reflecting the Previous Utterance of the Other Participant

AU - Yamamoto, Hirofumi

AU - Tanigaki, Koichi

AU - Sagisaka, Yoshinori

PY - 2004/1

Y1 - 2004/1

N2 - This paper proposes the construction of a language model for speech recognition in speech translation. The proposed language model is intended to improve speech recognition performance by taking account of the previous utterance content of the counterpart speaker. The previous utterance content of the counterpart speaker is represented by the intermediate language which is generally used by C-star (Spoken Language Translation Research Group). The proposed model is based on a language model that is dependent on the intermediate language representation. An experiment concerning the conversation of hotel reservations shows that the rate of incorrect word recognition is reduced by approximately 5.4% (from 14.7% to 13.9%), and the usefulness of the proposed model is demonstrated.

AB - This paper proposes the construction of a language model for speech recognition in speech translation. The proposed language model is intended to improve speech recognition performance by taking account of the previous utterance content of the counterpart speaker. The previous utterance content of the counterpart speaker is represented by the intermediate language which is generally used by C-star (Spoken Language Translation Research Group). The proposed model is based on a language model that is dependent on the intermediate language representation. An experiment concerning the conversation of hotel reservations shows that the rate of incorrect word recognition is reduced by approximately 5.4% (from 14.7% to 13.9%), and the usefulness of the proposed model is demonstrated.

KW - Conversation

KW - Intermediate language representation

KW - N-gram

KW - Statistical language model

KW - Task adaptation

UR - http://www.scopus.com/inward/record.url?scp=0347316435&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0347316435&partnerID=8YFLogxK

U2 - 10.1002/ecjb.10211

DO - 10.1002/ecjb.10211

M3 - Article

VL - 87

SP - 54

EP - 62

JO - Electronics and Communications in Japan, Part II: Electronics (English translation of Denshi Tsushin Gakkai Ronbunshi)

JF - Electronics and Communications in Japan, Part II: Electronics (English translation of Denshi Tsushin Gakkai Ronbunshi)

SN - 8756-663X

IS - 1

ER -