“Repurposing a Rhythm Accompaniment System for Pipe Organ Performance” by Nicholas Evans, Behzad Haki, and Sergi Jordà
In this paper, the authors used a traditional generative method that converts rhythm to harmony on multiple pipe organs enabled with MIDI. Rhythmic output, specifically onsets and velocities of notes, was generated by a Groove Transformer. This was fed into a Markov-based model in Max, which was trained during the performance from the live organists, and converts it into harmonies. The way it works is when the Groove Transformer generates a rhythmic pattern, each onset triggers the Markov model to generate the next pitch and duration to create an accompaniment. The accompaniment itself has multiple different voices that end up complementing the live organists.
“Looping slowly: Diffraction through the lens of nostalgia” by Benedict Gaster, Nathan Renney, and Jasmine Butt: The use of magnetic tape for a calm and slower approach as opposed to the fast-paced modern techniques of today