Influences of Integrating Dynamic Assessment into a Speech Recognition Learning Design to Support Students’ English Speaking Skills, Learning Anxiety and Cognitive Load

Abstract

Artificial intelligence (AI) technology has been progressively utilized in educational environments in recent years, due to the advances in computing and information processing techniques. The automatic speech recognition technique (ASR) provides students with instantaneous feedback and interactive oral practice for supporting a context with self-paced learning. Corrective feedback (CF) should be combined with ASR-based systems to enhance students’ speaking performance, and to reduce their cognitive load. However, learners’ perceptions of CF are mixed, and CF might give rise to learning anxiety. In this study, a dynamic assessment-based speech recognition (called DA-SR) learning system was designed to facilitate students’ English speaking. Moreover, a quasi-experiment was implemented to evaluate the effects of the proposed approach on students’ speaking learning effectiveness, via respectively providing the DA-SR and the corrective feedback-based speech recognition (called CF-SR) approaches for the experimental and control groups. The experimental results revealed that both the DA-SR group and the CF-SR group can effectively improve the students’ English speaking skills, and decrease their English speaking learning anxiety. Moreover, this study further demonstrated that the DA-SR approach successfully reduced students’ English class performance anxiety, and extraneous cognitive load in comparison with the CF-SR approach. It could be a valuable reference for designing English speaking learning activities in EFL learning environments.

Keywords

• artificial intelligence
• speech recognition
• corrective feedback
• dynamic assessment
• learning anxiety