By analyzing the importance of fingering in piano performance and the challenges posed by existing techniques, this study develops a 5G and AI-based piano fingering automation system. The system leverages deep learning and big data analysis to generate efficient and scientific fingering suggestions. Experiments demonstrate that the system can recommend correct fingering in 98% of cases, achieving 85% consistency with the fingering of professional pianists. The system effectively improves the efficiency and accuracy of students’ learning. After using the system for one week, 25 students experienced a 30% improvement in their playing skills and a 40% reduction in their error rate. The results indicate that the system holds significant potential for application in piano education at colleges and universities, effectively assisting in both learning and teaching.