This study investigates the effect of using the artificial intelligence (AI)-based application ELSA Speak on enhancing English speaking skills and reducing speaking anxiety among 25 junior eighth-grade students at SMP Negeri 32 Semarang, Indonesia. This study used a mixed-methods approach with a single-group pretest-posttest design. Quantitative data were collected through speaking tests and the Foreign Language Classroom Anxiety Scale (FLCAS) questionnaire and analyzed using the Wilcoxon Signed-Rank Test, while qualitative data were gathered through interviews and subjected to thematic analysis with the support of MAXQDA software. The results demonstrated a significant improvement in students' speaking skills from pre-to post-test, with a very large effect size and reduced speaking anxiety. Qualitative findings revealed that ELSA Speak enhanced students’ speaking abilities, particularly in pronunciation and vocabulary mastery, while effectively reducing speaking anxiety and fostering greater confidence in oral communication.
Copyrights © 2026