AI-Generated TikTok Microlearning Videos for Eight Grade EFL Learners’ Pronunciation of English Silent Letters

Abstract

This study investigates the use of AI-generated TikTok microlearning videos for teaching English silent letters to eighth-grade EFL learners. Many students experience difficulties in pronouncing silent letters because the relationship between English spelling and pronunciation is often inconsistent. To address this issue, the study employed AI-generated TikTok microlearning videos incorporating Natural Reader as an AI-based text-to-speech tool to provide pronunciation models of words containing silent letters. The study employed a quantitative pre-experimental design using a one-group pre-test and post-test approach involving 31 eighth-grade students at SMP Negeri 5 Tuban, Indonesia, during the 2025/2026 academic year. Data were collected through pronunciation pre-tests and post-tests and analyzed using descriptive statistics, the Shapiro-Wilk normality test, and the Wilcoxon Signed-Rank Test. The results showed an increase in the mean pronunciation score from 41.93 in the pre-test to 64.50 in the post-test. The Shapiro-Wilk test indicated that the data were not normally distributed (pre-test = 0.008; post-test = 0.009), therefore, the Wilcoxon Signed-Rank Test was employed and yielded a significance value of 0.000 (p < 0.05). The effect size was r = 0.865, indicating a strong effect. These findings indicate a statistically significant difference between students’ pronunciation performance before and after the use of the AI-generated TikTok microlearning videos. However, because the study employed a one-group pre-experimental design without a control group, the findings should be interpreted as evidence of improvement following the intervention rather than definitive evidence of causal effectiveness. The findings suggest that AI-generated TikTok microlearning videos may serve as a supplementary digital medium for teaching English silent letters to EFL learners