音声訂正:選択操作による効率的な誤り訂正が可能な音声入力インタフェース  [in Japanese] Speech Repair: Speech Input Interface Capable of Quick Error Correction by Using Selection Operation  [in Japanese]

Access this Article

Search this Article

Author(s)

Abstract

本論文では,ユーザが認識誤りを選択操作により効率的に訂正することが可能な「音声訂正」という音声入力インタフェース機能を提案する.音声訂正では,ユーザが音声入力を開始すると,認識結果を単語ごとに区切った表示と,区切られた各区間に対する他候補(競合候補)が発話の最中から次々と画面に描画され,ユーザは競合候補の中から本来の正解を選択するだけで認識誤りを訂正することが可能となる.また,音声訂正では,ユーザが発声中であっても訂正処理が可能な「即時誤り訂正機能」と,ユーザが意図的に発声を休止し,認識処理を一時中断させることが可能な「発話中休止機能」を実現する.25 人の被験者による評価実験を行ったところ,音声訂正は使いやすく,効果的な音声入力インタフェースであることが確認された.In this paper, we propose a speech input interface function, called "Speech Repair", which enables a user to easily correct recognition errors by selecting candidates. During the speech input, this function displays not only the typical speech-recognition result but also other competitive candidates. Each word in the result is separated by line segments and accompanied by other word candidates. A user who .nds a recognition error can simply select the correct word from the candidates for that temporal region. Furthermore, we introduce two additional functions: immediate correction function that enables the user to correct errors not only when the recognition process is complete but also whenever the user .nds erroneous words, and intentional suspension function that enables the user to intentionally suspend and resume the recognition process. Experimental results with twenty-five subjects showed that the speech-repair function is easy to use and effective interface.

In this paper, we propose a speech input interface function, called "Speech Repair", which enables a user to easily correct recognition errors by selecting candidates. During the speech input, this function displays not only the typical speech-recognition result but also other competitive candidates. Each word in the result is separated by line segments and accompanied by other word candidates. A user who finds a recognition error can simply select the correct word from the candidates for that temporal region. Furthermore, we introduce two additional functions: immediate correction function that enables the user to correct errors not only when the recognition process is complete but also whenever the user finds erroneous words, and intentional suspension function that enables the user to intentionally suspend and resume the recognition process. Experimental results with twenty-five subjects showed that the speech-repair function is easy to use and effective interface.

Journal

  • IPSJ journal

    IPSJ journal 48(1), 375-385, 2007-01-15

    Information Processing Society of Japan (IPSJ)

References:  14

Cited by:  10

Codes

  • NII Article ID (NAID)
    110006152211
  • NII NACSIS-CAT ID (NCID)
    AN00116647
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    1882-7764
  • NDL Article ID
    8649997
  • NDL Source Classification
    ZM13(科学技術--科学技術一般--データ処理・計算機)
  • NDL Call No.
    Z14-741
  • Data Source
    CJP  CJPref  NDL  NII-ELS  IPSJ 
Page Top