Automated Region-of-Origin Classification of Traditional Music from Peninsular Spain

Pablo Stutt Castro


Supervised by Andrew Jones; Moderated by Stuart M Allen

In recent years, interest in the field of Music Information Retrieval (MIR) has grown significantly, particularly in the field of Music Genre Classification (MCR). However, these studies often focus on contemporary musical genres, and traditional music is entirely absent from relevant datasets such as the GTZAN dataset. This project attempts to carry out some initial research in the field of automated classification of audio recordings of traditional music, and to research the challenges that this field presents. Due to the incredibly wide scope of this problem, this project focuses specifically on traditional music from peninsular Spain. A dataset will be built that is represenative of the different musical traditions of Spain. A model that can classify audio recordings into their region of origin will be trained.

Initial Plan (07/02/2022) [Zip Archive]

Final Report (27/05/2022) [Zip Archive]

Publication Form