Users’ perceptions of ASR as a writing tool: an analysis using the technology acceptance model
Danial Mehdipour-Kolour, Walcir Cardoso · Computer Assisted Language Learning · 2025
Automatic speech recognition (ASR) is a technology that automatically converts speech to text. While ASR has been widely studied in computer-assisted pronunciation training research, research on ASR in second language (L2) writing has not received the attention it deserves. Indeed, there are only a handful of studies employing ASR as a transcription mode for native and non-native English writers without learning disabilities. Therefore, to gain insight into whether writers accept ASR as a transcription mode, this study adopts a one-sample design involving 60 participants (native and nonnative English writers) to investigate their perceptions of using this technology for writing. Using the technology acceptance model 2, this study surveys participants after they engaged in ASR-based writing. The findings reveal that writers mostly accepted or were neutral towards using ASR for writing, and that there was no significant difference between native and non-native English writers in perceiving ASR as an acceptable mode of transcription. Implications and recommendations for future research are discussed.