Audio Based Music Genre Classification using Convolutional Neural Networks Sequential Model

Abstract

Music recommendation system has high demand in the modern world, and these types of systems recommend music according to music genres. The music combines lyrics and beautiful sounds, and lyrics can be in any language. Thus, sound plays an essential role in defining the genres of all music written in different. Therefore, the recognition of music genres is a challenging task. However, various studies have been found in this area that performed well. However, the performance of the system can be further increased. Thus, the proposed research aims to identify the music genres with good performance. This study used GTZAN Dataset, which is a publicly available dataset. This dataset has ten classes and around 1000 samples for each category. This system used the Convolutional neural network (CNN) model to recognize the performance of the music genres recognition system. The CNN model has ten layers, eight dense and two dropout layers. The achieved accuracy by the proposed model is 98.30%.