RETAIN (RESEARCH ON ENGLISH LANGUAGE TEACHING IN INDONESIA)
Vol. 13 No. 02 (2025)

The Use of Pronunciation Training Based on Automatic Speech Recognition of ELSA Speak Application for EFL Students

Febrianti, Rizky Liliani (Unknown)



Article Info

Publish Date
30 Jul 2025

Abstract

English is the most used language for communication in fields such as international trade, diplomacy, education, and international telecommunications. Knowing the importance of communication in this era, English is being taught worldwide, including in Indonesia, with the aim of equipping EFL students with proficiency to open up a wide range of future career opportunities. But, is not that easy to learn pronunciation, when practicing pronunciation, errors in phonology or pronunciation are common. Fortunately, technological advancements in education offer solutions for this case. The ELSA Speak application supported with Automatic Speech Recognition (ASR), helps students in practicing their pronunciation. A high school in Krian has adopted this application as a learning tool in English classes to improve the pronunciation skills of EFL students. This research aimed to discover how the implementation of the ELSA Speak application as student’s learning partner, the impact of the app on students’ pronunciation ability, and the students’ responses toward the performance of the application. This study used a mixed method that combined qualitative and quantitative data. The number of samples was 30 students. The findings revealed that: (1) The ELSA Speak was implemented as a partner for training difficult and new vocabulary pronunciation. This application offers flexible online pronunciation exercise, artificial speech recognition, pronunciation analysis and constructive feedback to improve the students’ speaking skills. (2) The average of students’ pre-test and post-test scores had growth up to 72%, the paired sample t-test using SPSS showed that the t-value > t-table (35.288 > 1.69913) with a significance level of 0.05 > 0.000. It can be concluded that H0 is rejected which means there is a difference between the pre-test and post-test results, and the result of effect size formula showed that the effect of this treatment was 0.64. The effect size between 0.2 < ES < 0.8 indicated to be a medium-category effect. (3) All the questionnaires related to the ELSA Speak application’s performance obtained positive responses, which means that the students were satisfied with the ELSA Speak application performance. In conclusion, the ELSA application proved its effectiveness as a tool and practice partner in improving students' pronunciation skills, the features of this app fill the needs of students to practice important aspects such as accuracy, fluency, stressing, and intonation to be able to pronounce words correctly.

Copyrights © 2025






Journal Info

Abbrev

retain

Publisher

Subject

Education Languange, Linguistic, Communication & Media

Description

RETAIN publishes articles within the scope of English Language Teaching and Applied Linguistics. RETAIN publishes articles within the scope of English Language Teaching and Applied ...