Human-Machine Interaction (HMI) technology has revolutionized the landscape of oral English education, offering new possibilities for improving learning efficiency and experiences. This paper presents an innovative teaching system that integrates real-time speech recognition and feedback capabilities with advanced natural language processing (NLP) and machine learning algorithms. The system is designed to provide personalized learning paths based on learners' performance data, ensuring tailored resources and guidance. Emphasizing user experience and interactive design, it aims to stimulate learner interest and motivation. Research findings indicate significant improvements in students' pronunciation, fluency, and grammar, alongside high levels of user satisfaction. However, challenges remain in fully replicating genuine human interactions and addressing technical limitations. Future work will focus on enhancing conversational abilities, personalization, and multimodal feedback mechanisms to better prepare students for real-world communication scenarios.