A methodology for sound scene manipulation based on the ray space transform

In this paper we devise a methodology for analysing and subsequently manipulating a sound scene acquired by means of a uniform linear array of microphones. The array signal is transformed in the ray space, i.e. a domain where acoustic rays are points; here we extract the source position and orientation in space and its radiation pattern, while its signal is extracted by a near-field beamformer. These descriptors can be easily manipulated and provided to any parametric rendering system. Through simulations we have proven the capability of the proposed method to perform different manipulations.