The Effects of Voice-based AI Chatbot on Korean EFL Students' Speaking Ability and Affective Factors

This study explores effects of voice-based AI chatbot on Korean EFL students' speaking ability and their affective factors: interest, belief, motivation, and anxiety. Participants were 44 freshmen students at a middle school in Korea. They were divided into one experimental group and one control group. During the ten-week experimental period, participants engaged in 10 chat sessions with the voice-based AI chatbot 'Echodot'. To take an examination of effects on the students' speaking ability, they took the NEAT speaking test as pre-and posttests. As for the affective factors, structured questionnaire-based surveys were conducted before and after the treatment to find out if there are any changes in affective factors. Findings reveal that the AI chatbot effectively contributes to improvement of speaking ability among EFL students. An analysis of survey results indicates affective aspects of AI chatbotassisted English learning changed positively over time. Accordingly, this study provides valuable insight into the use of AI chatbot in English learning in EFL contexts, suggesting that EFL teachers should try to integrate AI chatbot technology in their classrooms.