Extending the closed form image source model for source directivity
* Presenting author
The image source method (ISM) is a widespread tool in room acoustic simulation for the generation of early reflections, which can take into account the source and receiver directivity, as well as the absorption properties of the acoustic environment. A downside of the ISM is the increase in computational effort with increasing image source order caused by costly visibility checks that need to be done separately for each image source. In the special case of a shoebox shaped room, a closed form ISM exists, thus eliminating the need of visibility checks. However, the closed form ISM inherently assumes omni-directional sources, an assumption that is violated in almost all real-life scenarios. In the current study, we extend the closed form ISM for the source directivity. Moreover, we show that a closed form ISM combined with a stochastic model for the late reflections can render plausible binaural auralizations, even for non shoebox-shaped rooms. This makes it an appealing approach for applications that require perceptually plausible and computationally effective simulations rather than high physical accuracy.