Year: 2024
Type: article
Study design: Clinical Trial
Abstract: This study introduces ALL-Talk, a web-based autonomous learning platform designed to enhance English-speaking skills among EFL students. Informed by extensive literature on second language speech influences, speaking anxiety, corrective feedback, and technology integration in language learning, ALL-Talk leverages Microsoft Azure’s capabilities, including Text-to-Speech, Automatic Speech Recognition, Automatic Pronunciation Assessment, and immediate visual feedback mechanisms. ALL-Talk was evaluated over ten weeks with 17 EFL undergraduate students, focusing on enhancing their Business English communicative skills through improved fluency and pronunciation. Although changes in fluency between the pre-test and post-test were not statistically significant, t(16) = 1.29, p = .215, 95% CI [-2.19, 9.01], d = 0.31, male students improved significantly in overall pronunciation accuracy, t(5) = 3.19, p = .024, 95% CI [1.61, 15.06], d = 1.30. Additionally, both genders improved significantly in pronouncing /dʒ/, /z/, and /θ/. Preliminary evaluation and feedback indicate potential for ALL-Talk to support autonomous learning and speaking improvement in EFL contexts. However, future research should incorporate a longer evaluation period to yield more substantial research outcomes.
Source: ABAC Journal
Author: Simon Moxon
Institution: Walailak University
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: bronze
Funder: Walailak University
OpenAlex ID: W4404868973