The use of generative AI to improve the speaking skills of Chinese EFL learners in the vocational education context

Abstract In recent years, generative artificial intelligence (AI) has attracted increasing attention for its educational potential, particularly in English as a Foreign Language (EFL) contexts, due to its ability to simulate real-time interactions. This study investigates the effectiveness of generative AI in enhancing English speaking proficiency among Chinese vocational students, as well as their perceptions of using generative AI chatbots. A total of 54 Chinese vocational students participated in this mixed-methods study. Quantitative data were collected through pre- and post-speaking tests and a structured questionnaire, while qualitative data were obtained via semi-structured interviews. The findings revealed significant improvements in participants’ speaking performance across multiple scoring dimensions after the intervention. Questionnaire results indicated that most participants perceived generative AI as useful in enhancing speaking proficiency and learning motivation. Interview data further highlighted both the perceived benefits and practical challenges of AI use, including inaccurate speech recognition, unreliable information sources, and difficulties in constructing effective prompts. Overall, the findings suggest that generative AI can significantly enhance English speaking among Chinese vocational students and demonstrate its potential to support autonomous language learning.