Home > About Us > Technology > Technical Report > Technical Report No.22 2013

Communication Visualizing Technology "Voistrap®"

This report presents a technology called "Voistrap®" that visualizes face-to-face oral communications by sensing voices. This system employs ID card-size client devices on neck straps, wireless base stations to receive data, and a data server connected to the base stations via a LAN. Each client device has multiple small microphones embedded along its strap. When something is spoken in a setting like a meeting where each participant wears the device, the speaker is identified by the ratio of sound pressure levels between these microphones. To alleviate concerns about privacy and security, the system only senses the timing of speech, but not the contents thereof. Communication is analyzed by using such data as the amount of speech produced by each participant, and how each participant interacted with others in conversations according to the records on when each participant spoke. Verification experiments in meetings and training sessions confirmed that Voistrap® can visualize communication, and could be used to help people communicate better.

Author

  • Haruo Harada
    System Technology Laboratory, Research & Technology Group
  • Hirohito Yoneyama
    System Technology Laboratory, Research & Technology Group
  • Kei Shimotani
    System Technology Laboratory, Research & Technology Group
  • Akira Fujii
    System Technology Laboratory, Research & Technology Group
  • Yohei Nishino
    System Technology Laboratory, Research & Technology Group
  • Kiyoshi Iida
    System Technology Laboratory, Research & Technology Group