2023 Fiscal Year Annual Research Report
Development of an open-source speaking evaluation platform with automated scoring
| Project/Area Number |
20K00807
|
|---|
| Research Institution | Kochi University of Technology |
|---|
Principal Investigator |
ダニエルズ ポール 高知工科大学, 共通教育教室, 教授 (50307245)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
熊井 信弘 学習院大学, 付置研究所, 教授 (00248999)
|
|---|
| Project Period (FY) |
2020-04-01 – 2024-03-31
|
| Keywords | AI / EFL / automated essay scoring / language / grammar / listening / writing |
|---|
| Outline of Annual Research Achievements |
The acquisition of linguistic capabilities in a foreign language consists of several dimensions; one of the initial, yet critical, stages involves the learners' ability to discern and comprehend words and phrases during the act of listening attentively. This foundational skill sets the stage for the subsequent development of speaking competencies in the target language.
To facilitate learners in navigating and assimilating the rapid pace and complexity inherent in the auditory streams of a foreign language, a dictation tool has been innovatively designed and integrated as a plug-in within the Moodle learning management system. This technological intervention seeks to enhance learners' auditory processing and transcription abilities, thereby serving as a foundational tool in their language acquisition.
In addition to the development of listening tools, our previous speaking and writing activities were updated so that writing and speaking tasks could be automatically scored using freely available generative AI tools. The focus this year was on evaluating these productive tools. A qualitative study was conducted to compare the grammar scores generated by human raters against those generated by freely available generative AI tools using 152 English as a Foreign Language (EFL) student writing samples. The main objective of the study was to evaluate the effectiveness of generative AI in scoring the grammar and providing writing feedback. The study reveals the potential advantages of generative AI through the demonstration of automated essay scoring (AES) in an EFL setting.
|
