ASR-Based Practice: Exploring EFL Learners’ Experiences in Developing Pronunciation and Speaking Confidence

Abstract

Pronunciation accuracy remains a persistent challenge for EFL learners, particularly in contexts with limited opportunities for acquiring individualized feedback. Although ASR technology has been integrated into language learning platforms, learners’ subjective experiences of ASR-supported practice require further investigation, especially at Indonesian EFL university levels. The study explores EFL learners’ experiences of using Automatic Speech Recognition (ASR) features in English Central for developing pronunciation and speaking confidence. Using a descriptive qualitative design and semi-structured interviews, the data were analyzed through inductive content analysis to identify experiences of using ASR technology in developing pronunciation and speaking confidence, also addressing challenges in ASR use. The findings show that ASR-based practice supported pronunciation development through repetitive imitation, individual learning strategy, immediate and visual feedback, and also contributed to speaking confidence through creating a low-anxiety environment and conversational activity. However, learners reported challenges related to occasional misrecognition, consistency, also technical and environmental issues. The study suggests that ASR-based practice can be pedagogically integrated into EFL instruction to support individualized learning for pronunciation and speaking confidence, with appropriate instructional guidance.