The assessment model of online vocal music teaching quality under the optimized DL model

Abstract

In recent years, art disciplines have flourished in response to the requirements of the times, and online vocal music courses have been offered by major colleges and universities. The quality of education in colleges and universities is a key indicator of the quality of a school's talents, and improving its quality is the most important part of modernizing education. In this study, a new education index system is constructed to address the problems of traditional online vocal teaching quality evaluation methods, and a vocal teaching quality evaluation model based on an adaptive variant Genetic Algorithm improved Back Propagation neural network is proposed. The model is combined with the established new index system and used in the evaluation of vocal music teaching quality. Comparing the performance between the model and the model constructed by the traditional algorithm, the experimental results showed that the convergence speed of the optimized model was improved by 79.32%. The model adaptation reached convergence in the 19th time, and the adaptation value was stabilized at 1.34, which can be seen that the improved model has a stronger adaptive ability and a faster convergence speed. In summary, the results predicted by the model basically coincide with the trend of the actual evaluation results, and the existence of the error is small, indicating that the model can achieve a more comprehensive and scientific evaluation of education quality.