Unsupervised analysis of similarities between musicians and musical genres using spectrograms

Since the early days of the information era, digital music has been becoming one of the most consumed types of media, introducing the need for content-based tools that can search, browse, and retrieve music. Here we describe a method that can quantify similarities between musical genres in an unsupervised fashion, and computes networks of similarities between different musicians or musical styles. The method works by converting each song to its 2D spectrogram, and then extracting a large set of 2883 2D numerical content descriptors. The descriptors are weighted by their informativeness, and then the similarities between the musical styles are measured using the weighted distances between the musical pieces of each pair of musicians or genres. The similarities between all pairs provide a similarity matrix, which is visualized by a phylogeny. Experiments using 23 well known musicians representing seven musical genres show that the algorithm was able to separate the artists into groups that are in agreement with their respective musical genres. The analysis was done in an unsupervised fashion, and without any human definition or annotation of the musical styles.