Toward Piano Teaching Evaluation Based on Neural Network

Abstract

With the rise of piano teaching in recent years, many people participated in the team of learning steel playing. However, expensive piano teaching fees and its unique one-to-one teaching model have caused piano education resources to be very short, so learning piano performance has become a very extravagant event. The factors affecting music performance are varying, and there are many types of their evaluation such as rhythm, expressiveness, music, and style grasp. The computer is used to simulate this evaluation process to essentially identify the mathematical relationship between factors affecting music performance and evaluation indicators. The use of computer multimedia software for piano teaching has become a feasible way to alleviate the contradiction. This paper discusses the implementation method of piano teaching software, the issues of computer piano teaching, the computer teaching as one-way knowledge, and the lack of interaction. The neural network (NN) model is used to evaluate the piano performance and simulate teachers to guide students through their exercise. The performance of the proposed system is tested for the piano music of “Ode to Joy,” which is different from the collection of NN training samples, and is delivered ten times by another piano teacher, student A (piano level 6), and student B (piano level 5).