2013 Fiscal Year Final Research Report
Research on speech processing interface to improve speech intelligibility for elderly people
| Project/Area Number |
23560459
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Communication/Network engineering
|
|---|
| Research Institution | Kyushu Institute of Technology |
|---|
Principal Investigator |
NAKATOH YOSHIHISA 九州工業大学, 工学(系)研究科(研究院), 教授 (10599955)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MITSUNORI Mizumachi 九州工業大学, 大学院工学研究院, 教授 (90380740)
|
|---|
| Project Period (FY) |
2011 – 2013
|
| Keywords | 高齢者支援 / 音声明瞭化 / 音声データベース / インタフェース |
|---|
| Research Abstract |
The purpose of this study is to analyze for acoustic features of elderly speech, to improve the intelligibility using the acoustic features. First, we were compared with the transition difference for each vowel in adult and elderly. As a result, we found that the absolute value of the transition difference increases on /a/ and /i/ that need to move mouth and tongue largely. Secondly, we were compared with the distribution of the five vowels in the F1-F2 plane in adult and elderly. As a result, we found that the distribution is reduced in the elderly speech. Finally, we propose a method to improve the speech intelligibility by the formant frequency shift. We tested the effect of improving the intelligibility. A result of listening experiments, the intelligibility could be improved on word that need to move mouth and tongue largely.
|
