A New Music Classification Method based on BP Neural Network

Abstract

Abstract In order to improve the music classification accuracy, a new music classification method is proposed based on BP neural network in this paper. Firstly, the features are extracted, and then they are input into the BP neural network to train and obtain the optimal music classifier, lastly, the music classifier is tested by four music genres. The results show that the proposed method has improved the music classification accuracy and reduced the training time compared with traditional classification methods. And more importantly, the proposed method gives a new direction for the research and application of BP neural network.