Scenery based Kansei music selection for car audio systems

Consideration of the impression of occupants is a fundamental requirement for automatic music selection in car audio systems. This paper discusses the implementation of a Kansei music selection system that automatically selects a suitable music for the scenery during a drive. First, Kansei evaluation experiments are conducted to investigate the relationship between the scenery from a car and the impression of occupants. Based on the results of the evaluation experiments, a Kansei model linking the scenery and the impression is constructed. A music selection unit is also proposed to select suitable music genres based on the impression. The system is finally implemented using the Kansei model and the music selection unit. After the implementation, a simulation is conducted to test the performance of the proposed system using example scenery samples. The results of the simulation demonstrate that the system has sufficient ability to propose a suitable music based on scenery.