Establishment and application of diversified piano teaching system in colleges and universities from the perspective of artificial intelligence

Abstract

A method is proposed here for the automatic generation of piano fingerings based on a temporal difference reinforcement learning model. For the special scene generated by piano fingering, this method adopts the temporal difference reinforcement learning algorithm of Q-learning as the learning model, which is concise and clear, with high learning speed. A set of applicable music and visual mapping rules are proposed, and based on the analysis of existing music visualization cases, a music visualization design strategy suitable for piano learning is proposed. Then, an application is designed to assist users in piano learning. The visualization design, intelligent sparring, and evaluation visualization design, enable piano learners to understand the characteristics and emotional expression of music in multiple dimensions during the learning process, and can also intuitively understand their performance. In addition, you can also use the sparring mode to interact with the community to improve your performance and aesthetic level further. On the problems existing in online piano sparring learning at this stage and how to make it play an active role in the overall development of people, some suggestions and thinking are put forward, and it is believed that better sparring requires the joint efforts of society, platforms, teachers, and parents. From the current point of view, integrating online and offline should be the most suitable way for students to learn piano. Still, in the future, with the continuous development of network technology and the continuous progress of human thinking and ability, this model will be further improved.

Keywords

• reinforcement learning
• time difference
• piano teaching
• intelligent system application