A deep learning-based intelligent curriculum system for enhancing public music education: a case study across three universities in Southwest China

Responding to national aesthetic education reforms, this study introduces a deep learning-driven platform to enhance public music education in Southwest China's universities. Utilizing LSTM and Transformer models, the system analyzes real-time student learning, predicts mastery trends, and delivers personalized feedback via a cloud-based interface. A semester-long experiment across Guizhou Minzu University, Guizhou University, and Xichang University compared three groups: traditional instruction, MOOC-based hybrid teaching, and AI-enhanced personalized learning. The AI group achieved 32% higher post-test mastery scores, with predictive models maintaining high accuracy (RMSE < 0.15). The platform supports adaptive assessments, intelligent feedback, and instructional decision-making, offering a scalable solution for AI integration in arts education, particularly in culturally diverse, data-scarce settings. This work informs policymakers and developers aiming to modernize aesthetic education through advanced computing.