Automatic Music Clustering using Audio Attributes

Abstract—Music brings people together, it allows us to experience the same emotions. Currently musical genre classification is done manually and requires even the trained human ear considerable effort. Therefore, clustering songs automatically and then drawing valuable insights from those clusters is an interesting problem and can add great value to music information retrieval systems. Most of the work in this field has involved extracting the audio content from audio files. This paper explores a novel technique to cluster songs based on Echonest Audio Attributes and K-Means algorithm. I experiment with different sets of attributes and genres. Most notably, I achieve 75-85% accuracy on a 4 genre-dataset of Classical, Rap, Metal and Acoustic songs. This result is better than results reported for human song clustering.