Assessing the Effectiveness of Text-to-Speech and Automatic Speech Recognition in Improving EFL Learner’s Pronunciation of Regular Past-ed

Abstract

In recent years, there has been a growing interest in computer-assisted pronunciation training (CAPT) tools, especially with the integration of speech synthesis and automatic speech recognition (ASR) technologies in language learning applications. This quasi-experimental study aims to assess the effectiveness of text-to-speech (TTS) and ASR in improving the pronunciation of English past -ed among A1 EFL Learners level. The study used a pretest-posttest control experimental group design to assess learners' pronunciation skills. A questionnaire was also conducted to evaluate participants' perceptions and attitudes towards the use of both tools. The results of the t-test analysis demonstrated significant improvements in the post-test scores of the experimental group compared to the control group, highlighting the positive impact of TTS and ASR technologies on learners' phonological awareness, perception, and production skills. These findings suggest that TTS and ASR have the potential to facilitate different stages of pronunciation development and can aid learners in acquiring desired pronunciation features. Additionally, the questionnaire responses revealed positive perceptions and high satisfaction with TTS and ASR technologies, affirming their effectiveness as valuable resources for pronunciation practice and self-assessment. This study underscores the pedagogical implications of integrating TTS and ASR technologies in pronunciation instruction, enhancing learners' pronunciation skills and expanding the learning environment beyond traditional classroom boundaries