Plecto: A Low-Level Interactive Genetic Algorithm for the Evolution of Audio

The creative potential of Genetic Algorithms (GAs) has been explored by many musicians who attempt to harness the unbound possibilities for creative search evident in nature. Within this paper, we investigate the possibility of using Continuous Time Recurrent Neural Networks (CTRNNs) as an evolvable...

Full description

Saved in:
Bibliographic Details
Published inEvolutionary and Biologically Inspired Music, Sound, Art and Design Vol. 9596; pp. 63 - 78
Main Authors Ianigro, Steffan, Bown, Oliver
Format Book Chapter
LanguageEnglish
Published Switzerland Springer International Publishing AG 2016
Springer International Publishing
SeriesLecture Notes in Computer Science
Subjects
Algorithms & data structures
Artificial intelligence
Evolution
Genetic Algorithm
Interaction design
Neural network
User interface design & usability
Online AccessGet full text
ISBN9783319310077
3319310070
ISSN0302-9743
1611-3349
DOI10.1007/978-3-319-31008-4_5

Cover

More Information
Summary:The creative potential of Genetic Algorithms (GAs) has been explored by many musicians who attempt to harness the unbound possibilities for creative search evident in nature. Within this paper, we investigate the possibility of using Continuous Time Recurrent Neural Networks (CTRNNs) as an evolvable low-level audio synthesis structure, affording users access to a vast creative search space of audio possibilities. Specifically, we explore some initial GA designs through the development of Plecto (see www.plecto.io), a creative tool that evolves CTRNNs for the discovery of audio. We have found that the evolution of CTRNNs offers some interesting prospects for audio exploration and present some design considerations for the implementation of such a system.
ISBN:9783319310077
3319310070
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-319-31008-4_5