Speaker
Description
This presentation reports on an ongoing project that develops AI-supported web applications to improve English as a Foreign Language (EFL) learners’ spoken and written summaries and opinions on news texts. Drawing on key EFL pedagogical frameworks, including Schmidt’s Noticing Hypothesis, Swain’s Output Hypothesis, and process-oriented approaches to writing and speaking—the system aims to create learning conditions that promote deeper attention to linguistic form, meaningful output production, and iterative revision through guided feedback.
Developed with Vercel’s v0 platform and Google AI Studio, the applications use OpenAI and Google Gemini APIs to analyze the content, organization, and linguistic accuracy of learner responses. The system delivers instant, individualized feedback that helps learners notice gaps between their intended meaning and their actual language use, encouraging self-correction and fostering more coherent and accurate summaries and opinions. By enabling learners to revise their output through multiple cycles of noticing, output, and feedback, the system supports a process-oriented learning model that can be difficult to implement consistently in large EFL classes. The long-term goal is to integrate these AI-supported feedback tools into Moodle plug-ins, expanding Moodle’s capacity to facilitate autonomous learning, promote reflective practice, and provide scalable formative assessment aligned with contemporary EFL pedagogy.
| 発表日の希望 / Preferred Day | 2月28日（土）/ February 28 Saturday |
|---|---|
| MAJ R&D Grant | いいえ |
