The Effect of Text-to-Speech AI on Iranian EFL Learners’ Pronunciation

This study investigates the impact of Artificial Intelligence Text-to-Speech (AI TTS) technology on English pronunciation learning among Iranian EFL learners. Employing a true experimental design with quantitative analysis, the research evaluates the effectiveness of AI TTS feedback compared to traditional teacher-led methods. Data were collected via standardized pre- and post-tests administered to experimental and control groups. An independent samples t-test confirmed the equivalence of the groups at baseline. Results indicate that AI TTS instruction significantly enhanced pronunciation performance in Iranian EFL learners, leading to higher gains in both segmental and suprasegmental features compared to traditional methods. Specifically, participants in the experimental group showed a statistically significant improvement from pre-test to post-test, supported by large effect sizes. These findings suggest important implications for educators and curriculum designers seeking to integrate AI-driven tools into pronunciation instruction, fostering more interactive and learner-centered approaches to language learning.

کلیدواژه ها:

AI Text-to-Speech(TTS) ، Pronunciation Learning ، Iranian EFL Learners ، Segmental Feature ، Suprasegmental Feature