Development of an Automatic Evaluation System of Reading-aloud Performances by EFL Japanese Learners Using the Latest Speech Information Technologies
| Project/Area Number |
25580135
|
|---|
| Research Category |
Grant-in-Aid for Challenging Exploratory Research
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Foreign language education
|
|---|
| Research Institution | Tokyo International University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MINEMATSU Nobuaki 東京大学, 大学院工学系研究科, 教授 (90273333)
SOMEYA Yasumasa 関西大学, 外国語学部, 教授 (40348454)
KAWAMURA Akemi 東京国際大学, 言語コミュニケーション学部, 教授 (30326996)
NISHIKAWA Megumi 東海大学, 外国語教育センター, 准教授 (10453705)
Husky Kay 東京国際大学, 商学部, 准教授 (50237955)
|
|---|
| Research Collaborator |
KATO Shuhei HOYAサービス株式会社, 音声ソリューション事業部CS技術, 研究員
FUJITA Masaya HOYAサービス株式会社, 音声ソリューション事業部CS営業グループ, マネージャー
KUNIKOSHI Aki McRoberts B. V., 研究員
|
|---|
| Project Period (FY) |
2013-04-01 – 2016-03-31
|
| Project Status |
Completed (Fiscal Year 2015)
|
| Budget Amount *help |
¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Fiscal Year 2014: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2013: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
|
|---|
| Keywords | 音読 / 自動評価 / 発音の正確性 / 流暢さ / 自動化 / GOP / ROU / CV / 英語 / 評価 / 発音 / 正確性 / 発話速度 / 熟達度 / 変動係数 / 総合的熟達度 / 英会話 |
|---|
| Outline of Final Research Achievements |
Reading aloud is difficult for L2 learners. However, repeatedly reading aloud the same passage enables learners to enhance their automatized processing and produce more fluent performances. A new CALL system has been developed which can assess reading aloud both in terms of accuracy and fluency. The former is evaluated by one of the speech information technologies named GOP (Goodness of Pronunciation). The latter is assessed by automatic calculation of ROU (Rate of Utterance): the number of syllables per second. This system also calculates coefficient of variation (CV) which indicates the degree of automaticity of reading aloud. The results of the experiments in this study showed that ROU was more valid than GOP as an indicator of automatization in reading aloud and that CV could be used to measure the degree of automaticity in reading aloud, because the CVs significantly decreased as the repetition time went up and the proficiency levels of the participants increased.
|
Report
(4 results)
Research Products
(20 results)
