Abstract:Acoustic source localization and tracking (ASLT) are always hot topics for their wide applications, such as binaural hearing aids, smart devices, and audio–video conference systems. Numerous ASLT algorithms have already been proposed, however only very limited open databases were available for evaluation their performance. To date, most of these existing databases were recorded indoors using only a limited number of microphones. To the best of our knowledge, it still lacks of available open databases to evaluate ASLT algorithms for outdoor applications using an intermediate-sized microphone array. This paper describes a new database for evaluation ASLT algorithms, where a planar array with 32 MEMS microphones were designed to record a great number of vehicle whistles in outdoor environments with the vertical distance about 8 m away from the center of this array to the road surface. Besides the detailed description of this database, an example using the steered response power-phase transform sound source localization method was provided to show that the vehicle whistle database has been successfully built. The vehicle whistle database and the demo code have been uploaded and available online.

A vehicle whistle database for evaluation of outdoor acoustic source localization and tracking using an intermediate-sized microphone array