Musical Instruments Sound Classification using GMM

Abstract

Classification is the task of assigning objects to one of several predefined categories. In today’s decade classifying the musical signal from large data is a major task; the proposed work classifies the music into their respective classes. In this paper, the sound of the musical instruments classified automatically from the musical signals. Mel frequency cepstral coefficient is used as a feature extractor and the machine learning model namely Gaussian Mixture Model is used for classification. This system tested in ten different classes of musical instrument sound from two different instrument families such as Woodwind and Brass instruments. In this proposed work, the result yields satisfactory accuracy in the classification of musical instruments sound.