Design of Music Art Teaching Quality Evaluation System Based on Deep Convolution Neural Network Algorithm

Art teaching is an art subject with strong practicality. Its teaching philosophy focuses on improving students’ musical knowledge, cultivating students’ musical singing ability and deepening students’ artistic aesthetic sentiment. Exploring a new teaching mode of music art major in colleges and universities has become an important issue that music educators in colleges and universities pay attention to. In this paper, based on the deep convolution neural network algorithm, the teaching quality of music art is further studied. The research shows that CNN (Convolutional Neural Network) is the first to achieve basic convergence. When the training rounds are 40, it has reached basic convergence, and the accuracy of the verification set can reach $65 \%$. And when the training rounds of AI (Artificial intelligence) are 70 and 60, the basic convergence of AI is achieved, and the accuracy of verification set is $58 \%$ and $60 \%$. It can be seen from the data in Figure 4 that the error rate of CNN is the lowest, followed by AI (Ant colony) and finally AC. From the above experimental results, it can be concluded that CNN has the best effect in evaluating the quality of music art teaching. The algorithm based on deep convolution neural network has a very important influence on the efficiency and quality of music art teaching. Therefore, music teachers should actively take effective measures to mobilize the atmosphere of music classroom learning, so as to achieve the purpose of improving the quality of music art teaching.