Tích hợp hệ thống đánh giá kĩ năng nói bằng trí tuệ nhân tạo nhằm nâng cao hiệu suất nói tiếng Anh của sinh viên tại một trường đại học ở Thái Nguyên

Tóm tắt

Speaking plays an important role in English as a Foreign Language (EFL) learning, yet it remains difficult to develop because learners often lack regular practice opportunities and personalized feedback. In large classes, teachers may also find it challenging to monitor and support individual students effectively. This study employed a quasi-experimental design combined with interviews to examine the impact of an AI-based speech evaluation system, namely Language Confidence, on EFL students’ speaking performance. The findings showed that the AI-supported group achieved greater improvement than the group receiving traditional teacher feedback, particularly in pronunciation and fluency. Interview data indicated that students valued the system’s immediate feedback, opportunities for repeated practice, and low-pressure speaking environment, while the instructor observed increased student confidence and participation. The findings suggest that AI-based speech evaluation can serve as an effective supplementary tool for teaching speaking skills, provided that it is integrated with clear pedagogical purposes and the teacher’s instructional guidance.