Improving Pakistani College Students’ Speaking Skills Using AI-Driven Linguistic Input: An Experimental Study

Abstract

The primary goal of teaching English to ESL students is to equip them with the linguistic skills necessary for effective communication in real-world situations. However, in Pakistan, English language education often fails to yield positive outcomes, particularly in developing speaking skills among college ESL learners, largely due to an examination system prioritising rote memorization over practical language use. A major contributing factor is insufficient linguistic input, which limits students' ability to produce meaningful linguistic output, such as speaking. Despite extensive research, improving speaking skills in ESL learners remains a challenge, although few innovative teaching methods have been widely adopted. This experimental study explores the impact of AI-based applications, specifically 'Readlee' and '@Voice Aloud Reader', as tools for enhancing linguistic input through reading and listening. Fifty college ESL students were randomly divided into control and experimental groups. The experimental group received regular reading and listening input through AI-powered apps for 32 weeks, while the control group only read from their textbooks without AI-powered apps. Pre- and post-test scores were analysed using SPSS and t-tests. The findings revealed that participants in the experimental group exhibited significantly improved speaking proficiency compared to the control group. This study highlights the potential of AI-driven tools in fostering the speaking skills of college ESL learners and recommends their integration into ESL instruction for more effective language learning.