Investigating the Effects of Using Chatbot AI in Teaching Pronunciation to EFL Learners at Nguyen Tat Thanh University

Abstract

This study explores the effectiveness of Chatbot AI in improving final consonant pronunciation among 106 A2-level EFL learners at Nguyen Tat Thanh University based on action research. Eight learners were purposefully selected for in-depth analysis of experiencing Chatbot AI based on stimulated recall interviews and thematic analysis. The other 98 participants engaged in quasi-experimental design, divided into a control group (n = 49) and an experimental group (n = 49). The analysis utilized independent samples t-tests to compare the pre- and post-intervention error rates for fricative and plosive omission and substitution. The thematic analysis of the interviews highlighted that the 08 learners appreciated the tool’s real-time, detailed, and personalized feedback. Meanwhile, the statistical results revealed significant reduction in the experimental group’s error rate after using Chatbot AI: fricative omission decreased by a mean difference of 13.12, plosive omission by 24.00, fricative substitution by 3.86, and plosive substitution by 13.65, compared to the control group. These findings suggest that Chatbot AI effectively reduce pronunciation errors across fricative and plosive final consonants, with potential benefits for EFL instruction when integrated with teacher guidance.