Development of a Generative AI-Driven Vocabulary Learning App and Educational Program for Simultaneous Learning of Japanese and English

Abstract

Vocabulary is the foundation of language proficiency, and acquiring a large vocabulary and the ability to use it effectively is crucial for improving language skills (Webb & Nation, 2017). However, it has been pointed out that Japanese university students lack sufficient vocabulary in both their native language, Japanese, and their second language, English.

With the advent of generative AI, tasks such as writing and translation are being automated, and the significance of language acquisition is being reevaluated. In addition to simply knowing vocabulary (receptive vocabulary), the ability to produce and respond quickly (productive vocabulary, fluency) is becoming increasingly important, and the ability to keep up with generative AI is required.

To address this issue, we developed a vocabulary learning material that allows Japanese native speakers to learn Japanese and English vocabulary in tandem, and conducted a pilot study with first-year university students at our university. We focused in particular on developing materials that enable learners to acquire a vibrant vocabulary with "breadth," "depth," and "fluency" through learning at the sentence and text level, rather than just at the word level.

In the pilot study, we used a story creation task that involved creating a story using three vocabulary words per trial, and used generative AI to analyze errors and provide feedback to each student. In this presentation, we will report the results of the pilot study and the questionnaire responses from the participants, and discuss the challenges.



Author Information
Misa Otsuka, Jissen Women's Junior College, Japan
Kaoru Mita, Jissen Women's University, Japan

Paper Information
Conference: SEACE2025
Stream: Foreign Languages Education & Applied Linguistics (including ESL/TESL/TEFL)

The full paper is not available for this title




Comments & Feedback

Place a comment using your LinkedIn profile

Comments

Share on activity feed

Powered by WP LinkPress

Share this Research

Posted by James Alexander Gordon