Tonality Estimation in Electronic Dance Music: A Computational and Musically Informed Examination

Dr. Ángel Faraldo

Thesis supervisor: Dr. Sergi Jordà.

21/03/2018

Thesis Brief Description:

​My ​thesis revolves around the task of computational key estimation in electronic dance music, upon which​ I have performed​ three interrelated operations. First, I attempt to detect possible misconceptions within the task, which is typically accomplished with a tonal vocabulary overly centred in Western classical tonality, reduced to a binary major/minor model which might not accomodate popular music styles. Second, I present a study of tonal practises in electronic dance music, developed hand in hand with the curation of a corpus of over 2,000 audio excerpts, including various subgenres and degrees of complexity. Based on this corpus, I propose the creation of more open-ended key labels, accounting for other modal practises and ambivalent tonal configurations. Last, I describe my own key finding methods, adapting existing models to the musical idiosyncrasies and tonal distributions of electronic dance music, with new statistical key profiles derived from the newly created corpus.

Experience as a PhD Student:

My 4 years as UPF were possibly amongst the most intense of my life.... Not only because of my research, which was in itself a great time consumer, but also because I had the opportunity to get very actively involved in the larger project my research belonged to (GiantSteps), meeting partners, musicians and researchers from all across Europe in numerous trips. Given the nature of my work in between music and engineering, my stay at UPF gave me the opportunity to showcase the outcomes of my team not only at specialised academic conferences, but also at 1st class music events such as Sonar, Redbull Music Academy or the Amsterdam Dance Event. Furthermore, in the past four years, I have met fellow researchers and doctorandi that will count among my very best friends for the rest of my life.