Bayesian inference in direction of arrival analysis using spherical microphone arrays
* Presenting author
One challenging problem in acoustical applications using microphone arrays is determining the directions of arrival (DoAs) of multiple sound events. This work applies a unified Bayesian framework to address this problem in situations involving potentially multiple concurrent sound sources using spherical microphone arrays. It presents a two-level inferential problem of sound source enumeration and direction of arrival estimation. To this end, analytical models based on spherical harmonics are used to predict experimental data derived from spherical harmonics microphone arrays. The prediction is evaluated against the measured data in order to select the simplest such model that can adequately match the experimental data, thereby estimating first the number of sources, then their DoA information. This paper presents the analytic models, the two levels of Bayesian formulation, and analysis results to demonstrate the potential usefulness of this model-based Bayesian analysis for complex sound environments with potentially multiple concurrent sources. This paper discusses the impact of microphone array configurations upon DoA resolution. It additionally estimates performance of the method based on experimental results for simultaneous multiple sound sources, particularly in noisy or reverberant environments.