Background. Speaking is a crucial component of English language acquisition, yet it remains one of the most challenging skills to master among EFL learners. Traditional teaching methods often fail to provide sufficient opportunities for oral practice and personalized feedback. The flipped classroom method has emerged as a promising pedagogical approach, and the integration of Artificial Intelligence (AI) offers the potential to enhance its effectiveness. Purpose. This study aims to examine the application of an AI-based flipped classroom method and its impact on improving English speaking skills, particularly in the areas of fluency, pronunciation, coherence, and accuracy among senior high school students. Method. A quasi-experimental design was employed involving 60 students divided into control and experimental groups. The experimental group was taught using AI-integrated flipped learning, incorporating tools such as speech recognition and automated feedback systems. Pretests and posttests were administered, and data were analyzed using descriptive statistics and independent t-tests. Results. Findings indicate that the experimental group outperformed the control group in all speaking components. The most significant improvements were observed in fluency and pronunciation, attributed to regular, individualized AI-guided speaking practice. The consistency of score gains across students suggests that the method is both effective and equitable. Conclusion. The AI-based flipped classroom approach offers a valuable instructional model that merges active learning with intelligent technology. It facilitates personalized, adaptive, and continuous speaking practice, thereby supporting accelerated language acquisition. Further studies are needed to test its scalability and long-term impact in diverse learning environments.