実音響環境に頑健な音声認識を目指して  [in Japanese] Towards Robust Speech Recognition in Real Acoustic Environments  [in Japanese]

Search this Article

Author(s)

Abstract

音声認識技術は,著しい進歩を遂げているが,実際の環境に適用し,実用的な性能を得るためには,未だ種々の多くの問題が残されている.実際の環境には,種々の雑音の存在,マイクロホンや伝送路に起因する伝達特性の違い,部屋の音響伝達特性の影響などがあり,音声認識性能の劣化をもたらす.本稿では,このような要因に対する研究の動向,各手法の適用法,現在進行中のプロジェクトなどを紹介する.

Although speech recognition technologies have achieved remarkable progress, there still remain many various problems for use in real acoustic environments. In real environments, the speech recognition system suffers performance degradation by various kinds of noise, channel mismatches by microphones and transmission channels, and room reverberations. This paper introduces an overview of many efforts and technologies to overcome those problems, and related current research projects.

Journal

  • IEICE technical report. Speech

    IEICE technical report. Speech 102(35), 31-36, 2002-04-19

    The Institute of Electronics, Information and Communication Engineers

References:  58

Cited by:  22

Codes

  • NII Article ID (NAID)
    110004023368
  • NII NACSIS-CAT ID (NCID)
    AN10013221
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    09135685
  • NDL Article ID
    6162547
  • NDL Source Classification
    ZN33(科学技術--電気工学・電気機械工業--電子工学・電気通信)
  • NDL Call No.
    Z16-940
  • Data Source
    CJP  CJPref  NDL  NII-ELS 
Page Top