The role that sound spatialization plays in improving performance in an interactive installation : study of the correlation between gesture and localization of sound sources in space

• Main Author: Franco, Diogo Leichsenring
• Publication Date: 2018
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10400.14/28062
• topic: Sound spatialization; Sound source localization; Sound; Gesture; Motion capture; Interactive installation; Espacialização sonora; Localização de fontes sonoras; Som; Gesto; Captação de movimento; Instalação interactiva

The main objective of this research work is to study the correlation between gesture and localization of sound sources in space within the framework of interactive installations, based on theories of hearing and gesture. We have therefore chosen the experimental method by developing an interactive installation with which we carry out three different experiments, in which a subject’s hand is tracked by a Microsoft Kinect depth camera (motion capture) and a deictic gesture is used to trigger recorded music sounds and identify their localization in the horizontal plane. Thus, we manipulate the direction of sound and we measure the percentage of correct perceptual sound source localizations resulting from the participant’s responses in an Inquiry Mode Questionnaire in comparison with the actual directions of the gesture and perceptual sound sources provided by software. Descriptive and inferential statistics is applied to the collected data. The main results show that it is easier to define the origin of sound and that auditory perception is more accurate when its incidence is frontal in the horizontal plane, just as sound source localization theory predicts. Whereas 86.1% of all volunteers consider that their gesture coincides with the origin of sound in experiment 1, in which the use of their gesture in a certain direction produces a sound from that direction, only 58.1% admit the same in experiment 3, in which the same gesture is used to identify the system-predetermined localization of a perceptual sound source in an angle of 260o around a subject. At least 55.9% of all participants do not perceive that their gesture cannot coincide with the origin of sound in experiment 2, since sound is produced from the opposite surround direction, which seems to demonstrate that, when sounds are produced frontally or from the back and a person has the task of controlling their motion with a deictic gesture at the same time, his or her ability to identify the origin of sound generally diminishes, in addition to the already well-known reduced ability to identify it when it is in the median plane, if the head is not rotated. We therefore conclude that there is a relatively high correlation between gesture and localization of sound sources in space, but this is not as perfect as it could be owing to the limitations of the human auditory system and to the natural dependence of head movement on gesture.