Integrating automatic speech recognition and automated writing evaluation to reduce speaking anxiety and enhance speaking competence among Chinese EFL learners

Abstract

This study investigates how the combined use of automatic speech recognition (ASR) and automated writing evaluation (AWE) influences Chinese English as a foreign language (EFL) learners’ speaking anxiety and speaking performance. Adopting a mixed-methods experimental design, the study compares an experimental group (EG) that utilizes ASR-AWE tools with a control group (CG) that follows traditional speaking instruction. The results revealed significant reductions in speaking anxiety, especially in learners with low self-confidence, fear of making mistakes and fear of being laughed at. In terms of competence, the EG demonstrated statistically significant improvements in pronunciation, grammatical accuracy and interactive communication. Qualitative findings indicated that learners valued ASR for its real-time feedback on pronunciation and AWE for improving sentence structure and grammatical refinement. The integration of ASR and AWE supported autonomous, iterative speaking practice, providing targeted linguistic input and facilitating learner engagement. These findings offer practical implications for incorporating ASR-AWE technologies into oral communication curricula in EFL contexts, particularly in promoting student confidence and fluency in technology-supported environments.