日本語ディクテーション基本ソフトウェア(97年度版) Japanese Dictation Toolkit -1997 version-

Search this Article

Author(s)

Abstract

The Japanese Dictation Toolkit has been designed and developed as a baseline platform for Japanese LVCSR (Large Vocabulary Continuous Speech Recognition). The platform consists of a standard recognition engine, Japanese phone models and Japanese statistical language models. We set up a variety of Japanese phone HMMs from a context-independent monophone to a triphone model of thousands of states. They are trained with ASJ (The Acoustical Society of Japan) databases. A lexicon and word N-gram (2-gram and 3-gram) models are constructed with a corpus of Mainichi newspaper. The recognition engine JULIUS is developed for evaluation of both acoustic and language models. As an integrated system of these modules, we have implemented a baseline 5,000-word dictation system and evaluated various components. The software repository is available to the public.

Journal

  • Journal of the Acoustical Society of Japan E

    Journal of the Acoustical Society of Japan E 20(3), 233-239, 1999-05

    Japan Acoustical Society of Japan

References:  9

Cited by:  14

Codes

  • NII Article ID (NAID)
    110003106206
  • NII NACSIS-CAT ID (NCID)
    AA00256597
  • Text Lang
    ENG
  • Article Type
    Journal Article
  • ISSN
    03882861
  • NDL Article ID
    4715226
  • NDL Source Classification
    ZM35(科学技術--物理学)
  • NDL Call No.
    Z53-X48
  • Data Source
    CJP  CJPref  NDL  NII-ELS  IR  NDL-Digital 
Page Top