The Efficiency of Interactive Differential Evolution in Creation of Sound Contents: In Comparison with Interactive Genetic Algorithm

The Efficiency of Interactive Differential Evolution in Creation of Sound Contents: In Comparison with Interactive Genetic Algorithm

Makoto Fukumoto (Fukuoka Institute of Technology, Fukuoka, Japan), Ryota Yamamoto (Fukuoka Institute of Technology, Fukuoka, Japan) and Shintaro Ogawa (Fukuoka Institute of Technology, Fukuoka, Japan)
Copyright: © 2013 |Pages: 12
DOI: 10.4018/ijsi.2013040102
OnDemand PDF Download:
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Interactive Evolutionary Computation (IEC) is known as an effective method to create media contents suited to user’s preference and objectives. As one of the methods, we have applied Differential Evolution (DE) as evolutionary algorithm in IEC. This study investigated the efficacy of Interactive Differential Evolution (IDE) in comparison with Interactive Genetic Algorithm (IGA). Two listening experiments were conducted to investigate the efficacy: experiment 1 as a creating experiment with IDE and IGA, experiment 2 as a re-evaluating experiment. Target of the creation was warning sign sounds. Eighteen subjects participated in both of the experiments. The result of the experiment 1 showed that IDE overcame IGA, and significant increase of fitness was only observed in IDE. The result of the experiment 2, higher fitness value was observed in IDE, however, the difference between the two conditions was not significant. Parts of the results showed a possibility of IDE to create media contents.
Article Preview
Top

Interactive Differential Evolution For Creating Sign Sounds

This section provides a fundamental scheme of DE and the IDE method creating sign sounds. Furthermore, an IDE system creating sign sounds constructed for experiments is described.

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 9: 4 Issues (2021): 2 Released, 2 Forthcoming
Volume 8: 4 Issues (2020)
Volume 7: 4 Issues (2019)
Volume 6: 4 Issues (2018)
Volume 5: 4 Issues (2017)
Volume 4: 4 Issues (2016)
Volume 3: 4 Issues (2015)
Volume 2: 4 Issues (2014)
Volume 1: 4 Issues (2013)
View Complete Journal Contents Listing