Using Speakerly AI to Improve Speaking Skills: An Experimental Study on EFL Learners

Authors

  • Muhammad Sahrir Aras STKIP Tomakaka Tiwikrama

Abstract

This study employs quantitative approach. This study used an experimental research design, specifically a quasi-experimental design, to reveal the effectiveness of Speakerly AI in improving students’ speaking skills at STKIP Tomakaka Tiwikrama. This study was conducted at STKIP Tomakaka Tiwikrama, Jl. Moh. Hatta, Pasangkayu Regency. The sample of study were the second semester students of English Education Department, in academic year 2024-2025, which was totally sixteen students. The sample was taken by using Purposive Sampling technique where eight students chosen as experimental group and eight students as control group. The treatment which integrating Speakerly AI for speaking practice in experimental group and traditional method in control group was carried out in four weeks. Data was collected through Likert scale and speaking test. The result showed that the t-observed value exceeded the t-critical value (12.452 >1.873), indicated that the Speakerly AI was significant in improving students' speaking skills. In addition, the overall students’ perception was 4.11 from 5.00, indicated that the students perceived Speakerly AI positively, especially at the usefulness aspect.

Downloads

Download data is not yet available.

Downloads

Published

01.07.2026

How to Cite

Muhammad Sahrir Aras. (2026). Using Speakerly AI to Improve Speaking Skills: An Experimental Study on EFL Learners. Edumaspul: Jurnal Pendidikan, 10(1), 504–509. Retrieved from https://jurnal2.unimen.cloud/index.php/edumaspul/article/view/281