Application of Deep Learning in Vocal Music Teaching

Abstract

This paper builds a cloud-oriented speech teaching system based on deep learning mode. A synthetic algorithm for decision-making based on depth and hierarchy knowledge is presented. This algorithm calculates the weight matrix of the user and marks to achieve personalized speech recommendations. The whole process of cloud music teaching has been realized. The identification performance of the recorded database is compared with that of the Berlin database. It is proved that the identification ability of this algorithm is greatly improved.