Method for simulating creativity to generate sound collages from documents on the web

摘要

To create algorithmic art with documents available on the internet, artists must discover strategies for organizing those documents. In this project I used a graph structure based on Melissa Schilling's model of cognitive insight to reorganize sounds on the web using aural and lexical relationships. I was then able to generate music with these graphs using several different activation strategies. In section one I introduce my goals for this project. In section two I review other approaches to this problem and art that has influenced my approach. In section three I demonstrate techniques for organizing and collaging sounds from freesound.org. Sounds can be organized in a graph structure by exploiting aural similarity relationships provided by freesound.org, and lexical relationships provided by wordnik.com. Music can then be generated from these graphs in a variety of ways. In section four I show how my software was inspired by theories of creativity. Specifically I show how my software is an illustration of Melissa Schilling's graph model of cognitive insight. In section five, I elaborate on the pieces I've generated for this dissertation using this software and several other novel sound generating programs.