構音障碍診断支援のための音声特徴量の統合化と可視化に関する検討  [in Japanese] Speech Feature Integration and Its Visualization for a Computer-Aided Diagnosis of the Dysarthric Speech  [in Japanese]

Search this Article

Author(s)

Abstract

専門家の聴覚印象に基づく構音障碍の診断やリハビリにおいて,障碍音声に関する客観的情報(音素の歪み/置換/省略など)を提供できる診断支援システムの開発を行っている.我々は複数の音響音声特徴量からなる複合パラメータを提案し,それらを可視化した統合化音声画像が有する言語情報伝達機能を確認した.一方,辞書ベースの不特定話者単語認識実験から認識パラメータとしての有効性も示された.これらの知見は,障碍音声の音素歪みの可視化・定量化処理や置換/省略を想定した認識処理を行うことで,診断支援に有効な情報を提供できる可能性を示唆している.本稿では,音素の置換/省略/歪みの観点から行ってきた音声特徴量の統合化による複合パラメータ利用について検討し,それらの可視化として新たに試作した診断支援ツール機能について述べる.

For diagnosis and rehabilitation based on the specialists' auditory images, we have developed a diagnostic support system which provides the objective information including phoneme distortion, replacement and omission. We proposed the composite parameter including various acoustic-phonetic features, and confirmed the effectiveness of speech visualization in transmission of the linguistic information. Also, the effectiveness as recognition parameters was shown in a dictionary-based speaker-independent word recognition. Those knowledge suggest possibilities of diagnostic support by visualizing and quantifying distortion or by recognizing words with replacement or omission. In this paper, we investigate use of the composite parameters integrating the speech features from a view point of phoneme distortion, replacement and omission. Also, we describe functions of a trial version of the diagnostic support system as speech visualization.

Journal

  • IEICE technical report. Speech

    IEICE technical report. Speech 111(471), 39-44, 2012-03-01

    The Institute of Electronics, Information and Communication Engineers

References:  10

Codes

  • NII Article ID (NAID)
    110009546223
  • NII NACSIS-CAT ID (NCID)
    AN10013221
  • Text Lang
    JPN
  • Article Type
    ART
  • ISSN
    0913-5685
  • NDL Article ID
    023569633
  • NDL Call No.
    Z16-940
  • Data Source
    CJP  NDL  NII-ELS 
Page Top