Machine Intelligence Enabled Parametric Speaker: An Invention Towards Experience Sound

Abstract

This work proposes to channelize the conventional longitudinal behavior of sound which diverges in every direction after its generation to a beam like behavior seen in transverse waves, which is already conceived in a device named as parametric speaker. On top of this, using technologies such as AI, self-learning, and decision-making capabilities can be endorsed in the hardware-based device, exhausting its full capabilities. The goal of the proposed work is to provide an invisible screening and privacy to the individuals in public and semi-public areas associated with verbal mode of communication. The underlying work focuses on using the unconventional behavior of sound, possessing beam like properties produced through Artificial means by interference and Superposition principle, applicable to sound which is constructed by Parametric Speaker also known as directional Speaker. The following work in this paper attempts to showcase the possibilities of using and harnessing this specific sound properties with sophisticated technologies such as ML and Deep Learning to create semi private areas in public spaces which can not only help people within different age groups (mostly elderly population) but can also help people who are handicapped (such as visually impared people who can only receive information through auditory channels), the scope of the proposed work is not only limited to impared people but, also to general population, with the objective to increase the conventional information transfer system into personalized data delivery service through voice and speech.