Identification of Music Genre using Convolutional Neural Network

Main Article Content

Agrawal Shreyash, S. P. Dhanure, P. P. Rathod, Challawar Ashay, Gadekar Pritesh


This paper proposes the implementation of a Convolutional Neural Network (CNN) to classify music into its respective genre. The digital entertainment industry is booming right now and people are mostly listening to music online these days. Music plays a key role in our lives. The quantity of music being released on internet platforms is huge. But to manually classify music files is a hectic task for human beings. There is also a good chance of error in case of classification done by humans. The increase amount of work in this field recently has brought a great demand for automatic music genre classification. This paper presents a deep learning approach using Convolutional Neural Network (CNN). CNN model is trained using GTZAN dataset with ten musical genres and spectrogram of each audio file. The model essentially uses a neural specification to classify the music file into 10 different genres. CNN has good ability to work with various musical patterns. CNN has been mostly used in recent approaches and it is more efficient than standard machine learning approaches.

Article Details