Study on foresight technology of eating disorder and the dysphagia onset by the analysis on the wave pattern of speech sound.

2016 Fiscal Year Final Research Report

• Project/Area Number: 26560265
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Medical systems
• Research Institution: Seinan Jo Gakuin University
• Principal Investigator: KAI Tatsuo 西南女学院大学, 保健福祉学部, 教授 (60331899)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 嚥下障害 / 診断 / 予知 / 音声解析

Outline of Final Research Achievements

We conducted a search for “voice sound markers” to diagnose and predict eating disorder/ dysphagia by speech sound analysis. Establishment of voice recording conditions, selection of voice analysis software, selection of appropriate analysis parameters, examination to eliminate or minimize various factors affecting analysis, selection of a speech "word" for analysis were done to set experiment conditions. The search was carried out on (1) analysis before and after applying the treatment method to instantly improve dysphagia, (2) analysis by sex and age, (3) analysis by sex and age using various corpus. As a result, we found that it is essential to develop technologies to standardize the individual differences of spectrogram density and to develop technologies to capture partial attenuation of spectrograms appearing as complex patterns as numerical values.

Free Research Field

音声学