1995 Fiscal Year Final Research Report Summary
Construction of diagnosis system and training system for production of Japanese speech sounds conveying emotions and attitudes
| Project/Area Number |
05451157
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Japanese language education
|
|---|
| Research Institution | HIROSHIMA UNIVERSITY |
|---|
Principal Investigator |
HOSODA Kazumasa Hiroshima Univ., Fac.of Education, Prof., 教育学部, 教授 (30032759)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MACHI Hiromitsu Hiroshima Univ., Fac.of Education, Assoc.Prof., 教育学部, 助教授 (10116668)
IMADA Shigeko Hiroshima Univ., Fac.of Education, Prof., 教育学部, 教授 (70052248)
FUKADA Shozo Hiroshima Univ., Fac.of Education, Lect., 教育学部, 講師 (50228863)
MIZUMACHI Isao Hiroshima Univ., Fac.of Education, Prof., 教育学部, 教授 (50056199)
|
|---|
| Project Period (FY) |
1993 – 1995
|
| Keywords | Teaching of Japanese sounds / Utterance Intentions / Emotional Expressions in speech / Speech synthesis / Hiroshima dialect / Kansei Engeneering / Fuzzy measure / Integral model |
|---|
| Research Abstract |
The present research includes developments of "diagnosis model" and "training model" with emphasis on intention, emotion and attitude conveyed in utterances. Our findings are as follows.
1.Analysis of synthetic speech showed that amplitude was important for joy and sadness while fundamental frequency was for anger and fear.
2.Incredulous questions of one-word are different from simple questions and neutral utterances in terms of duration, F0 and amplitude.
3.One-word production training with visual feedback method was applied to learners of Japanese and some considerations were made for practical teaching.
4.CAI courseware was made open to the public on WWW,internet.
5.Binnan speakers learning Japanese were found to have difficulty (1) producing the difference between surprise and suspicion and (2) understanding neutral utterances.
6.Hiroshima dialect was acoustically compared with standard Japanese resulting in the finding that "rising tone" in Hiroshima dialect was a factor contributing to the perceptual difference between them.
7.CAI course was produced for Japanese learners to practice geminated stops.
8.Computer-aided diagnosis system for Japanese (WIDIAS ; Word Image Diagnosis Kansei Engineering System) was constructed using Kansei Engineering and Fuzzy measure/Integral Model. This system is able to represent human like evaluation of word feeling through the intellinegnt reasoning system of the model.
|
-
-
-
-
-
-
[Publications] NISHIBATA,Chikako, ASADA,Kentaro, SHINOZAKI,Daishi, XU Ai-hong, TADA,Miyuki, HOSODA,Kazumasa: "Prosodic Features of Japanese a and e Sounds Conveyed by Affective Meanings of Joy, Surprise, Discouragement, and Suspicion" Nihongo Kyoikugakka Kiyo. (1996)
Description
「研究成果報告書概要(欧文)」より
