Reducing Speaking Anxiety and Enhancing Oral Performance through Google ASR: Evidence from an ESP Hospitality Course

Abstract

In alignment with Taiwan’s 2030 national bilingual policy, this study explored the potential of Google’s Automatic Speech Recognition (ASR) as a supplementary tool for English for Specific Purposes (ESP) instruction in a hospitality course. The research focused on two critical aspects of oral communication: reducing speaking anxiety and enhancing oral performance among technological university students. A total of 50 participants were randomly assigned to an experimental group that practiced with Google ASR and a control group that followed traditional practice. Both groups received identical instruction in hospitality English, including tasks such as making reservations, seating guests, and taking orders, but differed in their after-class practice methods. Data collection consisted of pre- and post-tests of oral performance, an English speaking anxiety questionnaire, a learning satisfaction survey, and semi-structured interviews. Quantitative results revealed that the ASR group achieved significantly higher oral performance, reported lower speaking anxiety, and expressed greater learning satisfaction compared with the control group. Qualitative findings further highlighted three key benefits: improved pronunciation awareness, reduced classroom anxiety through low-stakes practice, and enhanced confidence and motivation, with some learners transferring these gains to workplace communication. These findings provide empirical evidence that Google ASR can be an effective pedagogical tool for ESP contexts such as hospitality English, supporting both linguistic development and affective needs. The study underscores the value of integrating ASR into EFL instruction as a means to reduce speaking anxiety and enhance oral performance.