Integrating Artificial Intelligence Mediated Communications to Enhance EFL Learners' Transactional Speaking Skills at SMP Tahfidz Mutiara Al-Akbar

Limited opportunities for structured speaking practice and delayed corrective feedback remain persistent challenges in junior secondary EFL classrooms, often resulting in suboptimal development of transactional speaking skills. Addressing this gap, the present study investigated the effectiveness of Artificial Intelligence (AI)-mediated communication through SpeechAce in improving students’ transactional speaking performance. This study employed a pre-experimental one-group pre-test–post-test design involving eight junior secondary students. Students participated in six instructional sessions integrating AI-based automatic speech recognition (ASR) feedback. Speaking performance was assessed using analytic scoring criteria, and data were analyzed using descriptive statistics and a paired-samples t-test. The results indicated a substantial improvement in students’ performance. The mean pre-test score was 47.36 (SD = 4.11), which increased to 52.24 (SD = 4.31) in the post-test. The mean difference of −4.36250 was statistically significant, t(7) = −6.328, p < .001, with a 95% confidence interval ranging from −5.99278 to −2.73222. These findings demonstrate that AI-mediated speaking practice contributed to consistent performance gains across participants. In conclusion, AI-assisted communication platforms such as SpeechAce can function as an effective pedagogical augmentation tool to enhance transactional speaking competence in EFL contexts, particularly by providing immediate feedback and structured repetition opportunities.