Online multi-speaker tracking using multiple microphone arrays informed by auditory scene analysis

Tracking multiple speakers with microphone arrays is used for practical applications such as video conferencing. An important task is the integration of multiple arrays with correct associations of multiple concurrent speakers. A single-array tracking approach based on CASA is extended here to probabilistic tracking with multiple arrays in order to handle a varying number of moving speakers over time and assign the concurrent localizations of multiple sensors to the speakers. Tracking is done simultaneously in angular and Euclidean space. The effectiveness of the method is shown with recordings of real speakers in a reverberant conference room by evaluation on the publicly available AV16.3 corpus.