Click Here for
Track Your Paper
ISSN:2454-4116

International Journal of New Technology and Research

Impact Factor 3.953

(An ISO 9001:2008 Certified Online Journal)
India | Germany | France | Japan

Development of a Support System for Voice Outputs and Character Recordings in Communication via the “Mouth-shape Character” Method

( Volume 5 Issue 11,November 2019 ) OPEN ACCESS
Author(s):

Tomoya Miyasaka, Norio Kato, Mamiko Kamoshida, Keiko Kawashima, Toshiaki Tanaka

Abstract:

The "mouth-shape character" is one of the communication means of ALS (Amyotrophic lateral sclerosis) patients and caregivers. Although this means allows communication between the patient and the caregiver without using any tools, it is a burden on caregivers to utter speech sounds and record characters. The purpose of this study is to develop a support system for speech utterances and character recordings in communication via the mouth-shape character method. The prototype support system (proto-system) consisted mainly of an electric character board and a speech voice recognition module. Using the proto-system, the communication experiments were conducted by healthy subjects role-playing both the patient and the caregiver, and the system operability was evaluated. A control experiment was performed using the conventional mouth-shape character method and the results using these two methods were compared. Although the communication speed in the proto-system was slightly inferior to that of the mouth-shape character method, the accuracy of the confirmed character recordings by the former method was high, indicating the possibility of reducing the burden on caregivers. In addition, since speech recognition adapted to the caregiver's voice can be used as an input trigger, both hands of the caregiver remain free, and the potential for a reduction in the caregiving workload was obtained.

DOI DOI :

https://doi.org/10.31871/IJNTR.5.11.17

Paper Statistics:

Total View : 606 | Downloads : 597 | Page No: 26-31 |

Cite this Article:
Click here to get all Styles of Citation using DOI of the article.