Fostering Students’ Autonomous Learning English Pronunciation Using Mobile-Based Artificial Intelligence
TL;DR
ELSA’s accessibility, immediate feedback mechanisms, and self-paced learning structure demonstrate its viability as a supplementary tool in English language instruction and highlight the usefulness of mobile-based AI applications in addressing pronunciation development needs within higher education contexts.
Abstract
This study explores the potential of ELSA (English Language Speech Assistant), a mobile-based artificial intelligence application, to support English pronunciation development among university students in self-directed learning contexts. Using a one-group pre-test post-test experimental design, we recruited 28 second-year university students (mixed gender), who had previously completed a pronunciation practice course. Participants engaged with ELSA’s AI-powered pronunciation features throughout the intervention period. Pre-test and post-test assessments were administered to examine changes in pronunciation proficiency, with data analyzed using paired-sample t-tests. Results revealed a substantial increase in mean pronunciation scores from 55.54 (SD = 13.077) to 80.18 (SD = 10.583) following the intervention (t (27) = -15.36, p < 0.001, Cohen’s d = 2.90). These findings illustrate ELSA’s practical utility in supporting autonomous pronunciation improvement among university learners. The application’s accessibility, immediate feedback mechanisms, and self-paced learning structure demonstrate its viability as a supplementary tool in English language instruction. The observed gains highlight the usefulness of mobile-based AI applications in addressing pronunciation development needs within higher education contexts. Future research could explore sustained engagement patterns with ELSA, identify which specific features contribute most to learning outcomes, and examine its applicability across diverse proficiency levels and learning environments.