Designing AI-Enhanced Pronunciation Instruction: A Conceptual Framework Integrating Perception, Feedback, and Learner Autonomy

Abstract

The topic of pronunciation has historically received limited support within second language (L2) education. One reason for the lack of support for pronunciation within L2 education is that the traditional setting of L2 classrooms is not conducive to the type of feedback that pronunciation instructors require from students. Although there are a variety of AI technologies that have been introduced to L2 education that provide benefits to language learners, the integration of these technologies into pronunciation education has been lacking. The AI-Mediated Pronunciation Learning Loop (AIM-Loop) model was created to incorporate three essential components of pronunciation education: perception, AI feedback, and learner autonomy into a framework that can be utilized to continuously facilitate the pronunciation learning of L2 students. Through reviewing studies on AI and pronunciation published between 2015 and 2025, it becomes possible to identify the mechanisms that contribute to the success of pronunciation education that utilizes AI technologies and frameworks. The results of these studies reveal that the best results occur when these AI technologies are incorporated into pronunciation education in a way that ensures that each of these three components is addressed in each pronunciation learning cycle for L2 students. By proposing the AIM-Loop, researchers can utilize AI technologies to better inform pronunciation education frameworks, and pronunciation educators can utilize this model to better incorporate artificial intelligence into their classrooms and methods of pronunciation education for their L2 students.