• DocumentCode
    3480339
  • Title

    Spoken Dialogue System for Learning Braille

  • Author

    Araki, Masahiro ; Shibahara, Kana ; Mizukami, Yuko

  • Author_Institution
    Dept. of Inf. Sci., Kyoto Inst. of Technol., Kyoto, Japan
  • fYear
    2011
  • fDate
    18-22 July 2011
  • Firstpage
    152
  • Lastpage
    156
  • Abstract
    Learning Braille, patterns of raised dots representing characters for visually impaired individuals, requires the assistance of another person to help identify the correspondence between the Braille pattern and the character. A self-learning system using speech technology would help a learner master Braille. We developed a spoken dialogue system that allows visually impaired individuals to self learn Braille, thus, eliminating the need for assistance. Our system comprises a mechanical Braille display that converts a character code to a Braille pattern, speech recognizer, speech synthesizer and dialogue manager. The software of the learning system was constructed using a platform for communication support systems, for hearing and/or visually impaired individuals, that is based on our Multimodal Interaction system architecture. Participants reported positive impressions as a result of using the developed Braille learning system.
  • Keywords
    character recognition; handicapped aids; interactive systems; speech recognition; speech synthesis; unsupervised learning; Braille learning; communication support systems; dialogue manager; mechanical Braille display; multimodal interaction system architecture; raised dot pattern recognition; self-learning system; speech synthesizer; speech technology; spoken dialogue system; visually impaired individuals; Adaptation models; Computer architecture; Data models; Hardware; Learning systems; Speech; Speech recognition; multimodal dialog system; speech application;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Software and Applications Conference (COMPSAC), 2011 IEEE 35th Annual
  • Conference_Location
    Munich
  • ISSN
    0730-3157
  • Print_ISBN
    978-1-4577-0544-1
  • Electronic_ISBN
    0730-3157
  • Type

    conf

  • DOI
    10.1109/COMPSAC.2011.27
  • Filename
    6032336