Automated Speech Scoring of Dialogue Response by Japanese Learners of English as a Foreign Language
| Project/Area Number |
18K00742
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 02100:Foreign language education-related
|
|---|
| Research Institution | Saga University |
|---|
Principal Investigator |
Hayashi Yuko 佐賀大学, 教育学部, 准教授 (10649156)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
近藤 悠介 早稲田大学, グローバルエデュケーションセンター, 准教授 (80409739)
石井 雄隆 千葉大学, 教育学部, 准教授 (90756545)
|
|---|
| Project Period (FY) |
2018-04-01 – 2023-03-31
|
| Project Status |
Completed (Fiscal Year 2022)
|
| Budget Amount *help |
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2020: ¥2,340,000 (Direct Cost: ¥1,800,000、Indirect Cost: ¥540,000)
Fiscal Year 2019: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2018: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
|
|---|
| Keywords | 発話自動採点 / スピーキング / 外国語教育 / 談話完成タスク / 発話自動採点システム / 自動採点 / 英語発話自動採点システム / 第二言語習得 / 高大接続 / 英語スピーキング / スピーキング評価 / スピーキングテスト / Computer-Based Testing / CBT |
|---|
| Outline of Final Research Achievements |
Numerous theoretical and/or practical problems make it challenging for teachers, researchers and policymakers to assess speaking proficiency in contexts that reflect real conversational situations. Using automated speech scoring technologies is one way to alleviate these problems. This study set out to build and evaluate a new system for automatically assessing learners’ speech within the context of an oral Discourse Completion Task (DCT). The DCT was administered to 210 undergraduate students of intermediate English language proficiency. The results show that the exact agreement between human and machine scores was moderately high, 72 %. This value is comparable to the extant literature on automated speech scoring and could serve as a foundation upon which to explore the applicability of the system in classroom settings. Areas for improvement in prediction accuracy and possible ways of untilising the current system in pedagogical settings are discussed.
|
| Academic Significance and Societal Importance of the Research Achievements |
今日、音声認識技術に加えて表情やモーションキャプチャーの技術も搭載し、バーチャル・エージェント（virtual agent）と英語で学習を進める双方向（やり取り）型の学習支援システムの開発・運用が進んでいる。しかし、（「学習」ではなく）自動「採点」（＝「話す」能力の測定）において、継続・発展性のある文脈が用いられている例は皆無に近い。本研究では、今日の外国語教育で重視されている、文脈や目的に応じた情報の理解と伝達が求められる談話完成タスクを用いた自動採点のシステム構築・精査に取り組み、外国語教育における自動評価の運用の可能性について検討・提案を行える点で学術的・社会的意義に富むと言える。
|
Report
(6 results)
Research Products
(14 results)
