E-learning Approaches Coupled with Musical Intelligence: Impacts on EFL Learners’ Pronunciation[Latin Thesis]

Sepehr Sabouri

Computer-Assisted Language Learning (CALL), in general, and Computer-Assisted Pronunciation Training (CAPT), in particular, have been now around for a while. In fact, advances in technology have not only promoted language teaching in the past few decades but have also necessitated the use of technology to a degree that we can now testify to the normalization of CALL in the language classroom. However, pronunciation training, as one of the most challenging tasks of an English teacher, seems to lag behind in benefiting from the myriads of advantages offered by technology. Despite the controversies hovering over native-speakerism ideologies, mostly posed by rather recent critical stances in language teaching, including English as an International Language (EIL), the role of intelligible pronunciation seems indispensable. This importance is especially emphasized when we consider the emerging and growing need of non-native English speakers to orally communicate with technological tools and devices. On the other hand, despite the close affinity between articulation and music, based on several shared features, it is surprising how the potential of musical intelligence in pronunciation training is under-researched in the field. Therefore, the present study set out to examine the integration of musical intelligence training in CAPT using Artificial Intelligence (AI) and Automatic Speech Recognition (ASR). For this purpose, 93 English words from Oxford Word Skills: Advanced were combined with words’ melody through Ableton Live 10 Suite and Adobe Audition computer applications in a cross-fading manner to use as pronunciation training tasks during the treatment and without words’ melody for control group. A list of 30 words was examined in both treatment and control groups to evaluate language learners’ pronunciation improvement over 18 sessions. Learners were asked to listen to the pronunciations and repeat them as the sound faded into melodies. The pre-test and post-tests were analysed by Vosk Toolkit ASR for confidence (precision) in terms of percentages, recorded as learners’ pre- and post-test scores. Also, interviews were conducted to get a better insight toward learners’ attitudes, motivations, and potential merits and demerits of the method. The results presented an enhancement in both groups over training courses in pronunciation skills, although there was no significant difference between the treatment group and the control group in terms of EFL learners' pronunciation skills. Moreover, it was indicated through the qualitative phase that implementation of this method can have a positive impact on EFL learners in terms of attitude and motivation levels. Despite the possible shortcomings of the study, the results seem very promising regarding the integration of musical intelligence with CAPT in language classrooms, or better yet, in pronunciation self-practice on ever-growing digital learning platforms.