AI-SUPPORTED SYNTHESIS OF SHEPARD-RISSET FREQUENCY SETS
Implementation of the new sound synthesis and analysis method in music composition.
- Adam Lukawski
The prospective research would be a continuation of Adam’s Master’s thesis in which he defined a new method for synthesis and analysis of music compositions with the use of a new taxonomy of musical parameters - “Periodic Fractal of Shepard-Risset frequency sets”. The doctoral research built on these discoveries would continue to expand the method of sound and music notation synthesis in relation to the “Periodic Fractal” algorithm with the use of digital signal processing techniques. These will be subject to artificial intelligence and blockchain technologies in order to create a new computer program giving access to a new artistic de-centralised network in which musical compositions could be discovered on a tree of possibilities (rather than composed) and might generate immediate economical value serving as cryptocurrency by itself that could be exchanged and “performed” between the users of the network. During the study, the examined methodology would be implemented in Adam’s artistic practice to create a portfolio of new compositions. The study would also hopefully trigger the discussion about the future of music composition in the world quickly shifting from the Information Era towards the so-called Imagination Age.
Keywords: deep learning, music composition, shepard tones, software, sound synthesis