A Bibliometric Approach and Meta-Analysis of Effects of Automatic Speech Recognition on Second Language Learning

Theoretical Framework

It is necessary to improve the design of ASR technologies due to their failures in communication and language training. Due to limited time, space, human teachers, collaborators, and lack of timely feedback, a growing number of language learners felt it difficult to train their second language speaking skills. ASR technologies may be able to complement this regret by providing enormous practice opportunities, timely corrective feedback, and training pronunciation (Mroz, 2018). Despite the popularity of ASR technologies, difficulties arise, however, when an attempt is made to implement them in second language learning (Kim et al., 2024). For instance, second language learners may feel it difficult to communicate with an unemotional machine and fail to follow the robot-based pronunciation. ASR sometimes fails to perceive learners’ utterances, leading to failure in communication and language training (de Vries et al., 2015).