Creation of Sound Contents by Extended Interactive Evolutionary Computation Using Heart Rate Variability

Interactive Evolutionary Computation (IEC) is known as an efficient method to create media contents which are suitable for individual users. To reduce user´s fatigue in IEC operation, Takagi et al. proposed extended IEC that uses physiological indices as evaluate value. We have already proposed an extended IEC that creates sound contents using heart rate variability. In this study, we re-evaluate created melody with subjective evaluation and heart rate variability. Genetic Algorithm was employed as evolutionary algorithm in IEC. In creation process, High Frequency component in heart rate variability was directly used as evaluation value. In re-evaluate experiment, the subjects evaluated subjective relaxation feeling for representative melodies in 7-point scale. In addition, we measure heartbeat information while the subject listens to melody. As a result of this study, gradual increase in heart rate variability was observed in creating experiment. In the result of re-evaluating experiment, gradual increase in a High Frequency component was also observed. Meanwhile, gradual decrease in subjective relaxation feeling was observed.