computer-assisted pronunciation training based on musical intelligence

Despite the current popularity of computer-assisted language learning (call), in general, and computer-assisted pronunciation training (capt), in particular, 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 educational technology. despite the controversies hovering over native-speakerism, mostly posed by rather recent critical stances in language teaching, including english as an international language, the role of intelligible pronunciation seems indispensable, especially when it comes to the growing need of learners to orally communicate with technological 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, an intact class of 19 intermediate efl learners were exposed to 18 sessions of pronunciation training based on musical intelligence. to do so, the pronunciations of 93 words were inserted into ableton live 10 suite for extraction of their melodies. subsequently, the melodies were combined with word pronunciations in a cross-fading manner using adobe audition and the resulting sequences were used as pronunciation tasks during the treatment. 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 precision in terms of percentages, as learners’ pre- and post-test scores. wilcoxon signed rank test showed a significant improvement in learners’ pronunciations during the treatment. 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.