Capturing the dynamics of cellular automata, for the generation of synthetic persian music, using conditional restricted Boltzmann machines

Davis, Darryl N.; Arshi, Sahar

Abstract

© Springer International Publishing AG 2017. In this paper the generative and feature extracting powers of the family of Boltzmann Machines are employed in an algorithmic music composition system. Liquid Persian Music (LPM) system is an audio generator using cellular automata progressions as a creative core source. LPM provides an infrastructure for creating novel Dastgāh-like Persian music. Pattern matching rules extract features from the cellular automata sequences and populate the parameters of a Persian musical instrument synthesizer [1]. Applying restricted Boltzmann machines, and conditional restricted Boltzmann machines as two family members of Boltzmann machines provide new ways for interpreting the patterns emanating from the cellular automata. Conditional restricted Boltzmann machines are particularly employed for capturing the dynamics of cellular automata.

Citation

Davis, D. N., & Arshi, S. (2017). Capturing the dynamics of cellular automata, for the generation of synthetic persian music, using conditional restricted Boltzmann machines. In Artificial Intelligence XXXIV; Lecture Notes in Computer Science (72-86). Cambridge, UK: Springer Verlag. https://doi.org/10.1007/978-3-319-71078-5_6

• Conference Name: SGAI-AI 2017
• Start Date: Dec 12, 2017
• End Date: Dec 14, 2017
• Acceptance Date: Oct 30, 2017
• Online Publication Date: Nov 21, 2017
• Publication Date: 2017
• Deposit Date: Dec 11, 2017
• Publicly Available Date: Jan 18, 2018
• Print ISSN: 0302-9743
• Electronic ISSN: 1611-3349
• Publisher: Springer Verlag
• Volume: 10630 LNAI
• Pages: 72-86
• Series Title: Artificial Intelligence XXXIV, Volume 10630 of the Lecture Notes in Computer Science
• Series ISSN: 0302-9743
• Book Title: Artificial Intelligence XXXIV; Lecture Notes in Computer Science
• Chapter Number: Six
• ISBN: 9783319710778; 9783319710785
• DOI: https://doi.org/10.1007/978-3-319-71078-5_6
• Public URL: https://hull-repository.worktribe.com/output/499702
• Publisher URL: https://link.springer.com/chapter/10.1007/978-3-319-71078-5_6