AI Applications for EFL Learners: Enhancing Speaking Performance and Reducing Anxiety with Gender-Based Analysis

This study explores the effectiveness of AI-based applications in enhancing speaking performance and reducing speaking anxiety among EFL learners, with a focus on gender differences. It examined three main areas: the impact of AI tools on students' speaking anxiety, their effect on speaking performance, and how these outcomes vary by gender. A quantitative approach was used through a quasi-experimental design involving pre- and post-tests and questionnaires. The intervention lasted six weeks and included two intact second-year English major classes at an Indonesian university, with 46 participants—22 males and 24 females. Findings showed a statistically significant reduction in students' speaking anxiety and a notable improvement in speaking performance after using AI-based speaking platforms. These results suggest that AI tools are effective in supporting English-speaking practice for EFL learners. Regarding gender, female students exhibited slightly higher anxiety levels, though not significantly different from males. However, females performed significantly better in speaking tasks than their male peers. This indicates that gender may play a role in how learners engage with AI-assisted language learning tools. The findings have important implications for integrating AI in language education, particularly in fostering more equitable and effective speaking instruction. Nonetheless, due to the relatively small sample size and the use of a quasi-experimental design with only two intact classes, the generalizability of these results is limited. Future studies with larger and more diverse populations are recommended to validate and expand on these findings.