Musical Instrument Recognition based on Convolutional Neural Network

In order to effectively solve the problems of traditional musical instrument recognition, which only aims at the recognition of single tone musical instrument and main musical instrument of polyphonic music, and limited data set and effective classification features. This paper constructs a more efficient instrument classification and recognition system with the help of convolution neural network. Starting with the musical theory knowledge of musical instruments, this paper extracts the key characteristics of musical instruments more efficiently with the help of neural network algorithm, and constructs an order of magnitude feature classification model in line with cognitive logic. Through the model classification experiment, it is found that the average recognition accuracy of convolutional neural network model for musical instruments is 0.801 and that for pure musical instruments is 0.96, verified the effectiveness of the model designed in this article in instrument recognition.