• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2005 Fiscal Year Final Research Report Summary

Speech recognition accepting utterances including out-of-vocabularies

Research Project

Project/Area Number 14380168
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Research Field Intelligent informatics
Research InstitutionWaseda University

Principal Investigator

SAGISAKA Yoshinori  Waseda University, Graduate School of Global Information and Telecommunication Studies, Professor, 大学院・国際情報通信研究科, 教授 (70339737)

Co-Investigator(Kenkyū-buntansha) SHIRAI Katsuhiko  Waseda University, School of Science and Engineering, Professor, 理工学部, 教授 (10063702)
KOBAYASI Tsunori  Waseda University, School of Science and Engineering, Professor, 理工学部, 教授 (30162001)
YAWIMOTO Hirofumi  Advanced Telecommunications Research Institute International, Senior Researcher, 主任研究員 (00395013)
Project Period (FY) 2002 – 2005
Keywordsstatistical language model / out of vocabulary (OOV) / hierarchical language model / continuous speech recognition / task-free speech recognition
Research Abstract

A speech recognition scheme was studied to accept utterances including out-of-vocabularies (OOVs). A hierarchical statistical language model was newly proposed to cope with OOVs and speech recognition experiments have been carried out to confirm its effectiveness. In this language model, we described word-neighboring characteristics of unregistered expressions and constituent phonotactic constraints statistically independently to cope with unregistered expressions. The upper layer of this hierarchical model consists of inter-word statistics expressed by multi-dimensional composite word N-grams and the lower layer expresses infra-word statistical phonotactics using multi-dimensional composite sub-word units. A series of speech recognition experiments have shown that this language modeling enables the effective use of independent statistics and achieved high recognition performance for utterances including OOVs. By expandingthis lower layer model for single words such as personal names a … More nd city names to much longer named entity such as book titles and movie titles, we have successfully shown the validity of this modeling to other unregistered expressions consisting of multiple words. This success suggests that the proposed language model is effective for OOVs task independently and the possibility of a task-free statistical language model by integrating different statistical constraints independently.
In speech recognition experiments, long unregistered expressions for movie titles were expressed by multi-dimensionalcomposite word N-grams as a lower-layer model. Experimental results showed that the proposed model recognition accuracy almost corresponded to the theoretical upper limit obtained by registering all OOVs as recognition lexicons. Furthermore, multiple Markov models have been automatically obtained by splitting OOV characteristics into multiple lower layered models. The use of word-class intrinsic models and automatically derived unsupervised models were proved to be useful for general unspecified OOVs, which gives a guideline of building statistical language models according to the size and the quality of available language data. Less

  • Research Products

    (20 results)

All 2005 2004 2003 2002

All Journal Article (20 results)

  • [Journal Article] Speech recognition of a named entity2005

    • Author(s)
      Tasuhiko Tomita, Yoshiyuki Okimoto, Hirofumi Yamamoto, Yoshinori Sagisaka
    • Journal Title

      Proc. ICASSP2005 I

      Pages: 1057-1060

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Speech Recognition of 00V Expressions and Words2005

    • Author(s)
      Tetsuhiko Tomita, Yoshiyuki Okimoto, Hirofumi Yamamoto, Yoshinori Sagasaki
    • Journal Title

      Proc. SNLP2005 I

      Pages: 273-278

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] 未知固有表現を含む音声の認識2005

    • Author(s)
      富田達彦, 沖本純幸, 山本博史, 匂坂芳典
    • Journal Title

      情報処理学会研究報告

      Pages: 117-122

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] 未登録固有表現と未登録単語を含む音声の認識2005

    • Author(s)
      富田達彦, 沖本純幸, 山本博史, 匂坂芳典
    • Journal Title

      日本音響学会2005年秋季研究発表会講演論文集

      Pages: 45-46

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Speech recognition of a named entity2005

    • Author(s)
      Tatsuhiko Tomita, Yoshiyuki Okimoto, Hirofumi Yamamoto, Yoshinori Sagisaka
    • Journal Title

      Proc.ICASSP2005 I

      Pages: 1057-1060

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] Speech Recognition of OOV Expressions and Words2005

    • Author(s)
      Tatsuhiko Tomita, Yoshiyuki Okimoto, Hirofumi Yamamoto, Yoshinori Sagisaka
    • Journal Title

      Proc.SNLP2005 I

      Pages: 273-278

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] Speech recognition of unregistered expressions2005

    • Author(s)
      Tatsuhiko Tomita, Yoshiyuki Okimoto, Hirofumi Yamamoto, Yoshinori Sagisaka
    • Journal Title

      IPSJ SIG Technical Reports

      Pages: 117-122

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] Speech recognition of OOV expressions and OOV words2005

    • Author(s)
      Tatsuhiko Tomita, Yoshiyuki Okimoto, Hirofumi Yamamoto, Yoshinori Sagisaka
    • Journal Title

      2005 Autumn Meeting Acoustical Society of Japan

      Pages: 45-46

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] 複数のマルコフモデルを用いた階層化言語モデルによる未登録語録認識2004

    • Author(s)
      山本博史, 小窪浩明, 菊井玄一郎, 小川良彦, 匂坂芳典
    • Journal Title

      電子情報通信学会論文誌(D-II) 12

      Pages: 2104-2111

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Mis-recognized utterance detection using hierarchical language model2004

    • Author(s)
      Hirofumi Yamamoto, Genichiro Kikui, Yshinori Sagisaka
    • Journal Title

      Proc. ICSLP2004 (International Conference on Speech Processing)

      Pages: 1025-1028

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] 未知固有表現を含む音声の認識2004

    • Author(s)
      富田達彦, 沖本純幸, 山本博史, 匂坂芳典
    • Journal Title

      日本音響学会2004年秋季研究発表会講演論文集 I

      Pages: 59-60

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Out-of-Vocabulary Word Recognition with a Hierarchical Language Model Using Multiple Markov Model2004

    • Author(s)
      Hirofumi Yamamoto, Hiroaki Kokubo, Genichiro Kikui, Yoshihiko Ogawa, Yoshinori Sagisaka
    • Journal Title

      The Journal of The Institute of Electronics, Information and Communication Engineers Vol.87

      Pages: 2104-2111

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] Mis-recognized utterance detection using hierarchical language model2004

    • Author(s)
      Hirofumi Yamamoto, Genichiro Kikui, Yoshinori Sagisaka
    • Journal Title

      Proc.ICSLP2004

      Pages: 1025-1028

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] Speech recognition for unregistered expression of a class2004

    • Author(s)
      Tatsuhiko Tomita, Yoshiyuki Okimoto, Hirofumi Yamamoto, Yoshinori Sagisaka
    • Journal Title

      2004 Autumn Meeting Acoustical Society of Japan I

      Pages: 59-60

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] Word Class Modeling for Speach Recognition with Out-of-Task Words Using a Hierarchical Language Model2003

    • Author(s)
      Yoshihiko Ogawa, Hirofumi Yamamoto, Yoshinori Sagisaka
    • Journal Title

      Proc. Eurospeech2003

      Pages: 221-224

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] タスク外語彙のための構造化クラス言語モデル2003

    • Author(s)
      小川良彦, 山本博史, 匂坂芳典, 小窪浩明, 菊井玄一郎
    • Journal Title

      日本音響学会2003年秋季研究発表会講演論文集 I

      Pages: 83-84

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Word Class Modeling for Speech Recognition with Out-of-Task Words Using a Hierarchical Language Model2003

    • Author(s)
      Yoshihiko Ogawa, Hirofumi Yamamoto, Yoshinori Sagisaka
    • Journal Title

      Proc.Eurospeech2003

      Pages: 221-224

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] Word Class Modeling for Speech Recognition with Out-of-Task Words Using a Hierarchical Language Model2003

    • Author(s)
      Yoshihiko Ogawa, Hirofumi Yamamoto, Yoshinori Sagisaka, Hiroaki Kokubo, Genichiro Kikui
    • Journal Title

      2003 Autumn Meeting Acoustical Society of Japan I

      Pages: 83-84

    • Description
      「研究成果報告書概要(欧文)」より
  • [Journal Article] タスク外語彙を含む音声の認識2002

    • Author(s)
      小川良彦, 磯貝俊太郎, 匂坂芳典, 大西茂彦, 山本博史, 菊井玄一郎
    • Journal Title

      日本音響学会2002年秋季研究発表会講演論文集 I

      Pages: 143-144

    • Description
      「研究成果報告書概要(和文)」より
  • [Journal Article] Speech recognition for out of vocabularies2002

    • Author(s)
      Yoshihiko Ogawa, Shuntaro Isogai, Yoshinori Sagisaka, Shigehiko Onishi, Hirofumi Yamamoto, Genichiro Kikui
    • Journal Title

      2002 Autumn Meeting Acoustical Society of Japan I

      Pages: 143-144

    • Description
      「研究成果報告書概要(欧文)」より

URL: 

Published: 2007-12-13  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi