Expressive sonification of footstep sounds

In this study we present the evaluation of a model for the interactive sonification of footsteps. The sonification is achieved by means of specially designed sensored-shoes which control the expressive parameters of novel sound synthesis models capable of reproducing continuous auditory feedback for walking over grounds of different nature, such as muddy ground and iced snow. The sensored-shoes consist of footwear equipped with pressure sensors, small loudspeakers, and haptic actuators responding to foot movement with synthetic sonic and haptic feedback. In a previous study, sounds corresponding to different grounds were associated to different emotions and gender. In this study, we used an interactive sonification actuated by the sensored-shoes for providing auditory feedback to walkers. In a factorial design we asked subjects to walk (using the sensored-shoes) with four different emotional intentions (happy, sad, aggressive, tender) and for each emotion we manipulated the ground texture sound four times (wood panels, linoleum, muddy ground, and iced snow). Preliminary results show that walkers used a more active walking style (faster pace) when the sound of the walking surface was characterized by an higher spectral centroid (e.g. iced snow), and a less active style (slower pace) when the spectral centroid was low (e.g. muddy ground). Harder texture sounds lead to more aggressive walking patters while softer ones to more tender and sad walking styles. These results can be used for the design of new interactive floors where the sound feedback of footstep sounds can influence the behavior of users, and also in therapy application when the control of the walking style of a client is a desired goal.