運用生成式人工智慧之英語歌曲多模態創作於EFL教育

Utilizing Generative Artificial Intelligence-Based Digital Multimodal Creation of English Songs for EFL Education

在語言學習領域，研究指出音樂可提升學習成效，尤其是詞彙與語法學習。音樂與語言教學的結合，本質上即符合多模態教學的理念。近年來，生成式人工智慧（GAI）也在語言教學領域引起了廣泛的矚目。然而，基於生成式人工智慧之英語歌曲多模態創作（GAI-MC）於語言教與學的應用及相關研究仍相對缺乏，本研究運用基於GAI-MC之方法，在北部某大學英文系的「人工智慧在語言學習的運用」課程中，針對28位學生進行了為期六週的單組前、後測實驗設計，以評估GAI-MC對EFL學習者自我效能感及後設認知覺察的影響。本文說明自我決定理論的三大核心需求，包含自主性、勝任感與關聯感於GAI-MC的實踐中，並將理論與實踐相結合，以迎合GAI時代對英語教育的新需求。

In the field of English as a Foreign Language (EFL) education, using music as a pedagogical medium has been empirically proven to significantly enhance learners’ learning outcomes. The integration of music with language education inherently aligns with the multimodal teaching approach. Previous research has primarily focused on the benefits of listening to and singing songs on learners’ vocabulary and grammatical usage. Although Generative Artificial Intelligence (GAI) has garnered widespread attention in the realm of language teaching, studies on the application potential of GAI-based multimodal creation of English songs (GAI-MC) in language learning remain scarce. This study employed the GAIMC approach in a six-week single-group pretest-posttest experimental design (N = 28) at a university in northern Taiwan to assess the impact of GAI-MC on EFL learners’ self-efficacy and metacognitive awareness. The paper elucidates the practice of the three core needs of Self-Determination Theory– autonomy, competence, and relatedness– within the GAI-MC approach and demonstrates the integration of theory and practice to meet the new demands of English education in the contemporary GAI era.