Using ASR in language classrooms: How does it compare to human listeners? (A replication study)
Résumé
With the advent of artificial intelligence in education, researchers have explored the pedagogical potential of automatic speech recognition (ASR) programs for pronunciation instruction in foreign language classrooms. This study replicates Inceoglu et al.’s (2023) study that compares ASR outputs to human listener’s assessments of L2 English for intelligibility, comprehensibility, and error type. The overall results support the idea that current ASR technology may be particularly useful for lower proficiency learners, with some pedagogical provisos.