The popularity of piano lessons has increased recently, and many individuals are taking the instrument. Using Web technology as its backbone and high-performance streaming media technology, this platform for teaching piano playing can efficiently address the issues of a limited audience, a steep learning curve, and a shortage of available materials in today's online education landscape. This kind of instruction may help bring the classroom closer together. The goals of this study are to increase the effectiveness of communicating via online text, to make it easier for students to get expert guidance from teachers, to satisfy students' interest in learning, to elevate students' piano effectiveness and creation, and to promote the widespread use of the piano. Using a Multiple Linear Regression Algorithm (MLRA), this study makes recommendations for designing an intelligent piano-playing teaching system, looks into the implementation of the piano-teaching system, and offers a technique to evaluate piano-playing using the MLRA model for the difficulties of online piano training. Teaching the piano is greatly aided by this paper's portrayal of an instructor encouraging students to keep up with their practice. The study results stated that MLRA's recognition accuracy is 95% and its mean square error rate is 12%.