Self-Directed Use of Voice-Enabled AI Chatbots for Improving English Speaking: A Case Study with Doubao

This study investigates the effectiveness of self-directed use of AI chatbots, specifically Doubao, in improving English-speaking skills, as well as the challenges learners encounter and the strategies they find effective. Using qualitative and quantitative data from 40 participants collected through 3 surveys over two months, the results indicate that Doubao significantly enhanced learners' English-speaking abilities by increasing practice frequency and addressing challenges associated with traditional methods. Participants valued the chatbot's accessible, low-pressure environment, which encouraged spontaneous practice. However, technical limitations, user-related barriers, and interaction challenges hindered the maintenance of meaningful conversations. Effective strategies included taking initiative, incorporating practice into daily routines, customizing chatbot settings, and contextualizing practice with clear goals. These findings contribute to the growing body of research on AI-assisted language learning, offering practical insights for both learners and educators. Further research is recommended to address technical and interaction-related issues to maximize the chatbot's effectiveness and user experience.