A Content-Based Music Similarity Function
Beth Logan, Ariel Salomon · 2001
We present a method to compare songs based solely on their audio content. Our technique forms a signature for each song based on K-means clustering of spectral features. The signatures can then be compared using the Earth Mover's Distance [14] which allows comparison of histograms with disparate bins. Preliminary objective and subjective results on a database of over 8000 songs are encouraging. For 20 songs judged by two users, on average 2.5 out of the top 5 songs returned were judged similar. We also found that our measure is robust to simple corruption of the audio signal and that meaningful visualizations of the data are possible using this similarity measure.