Automatic Switching Between Different Cameras at a Video Conference Endpoint Based on Audio

A video conference endpoint includes predefined main and side audio search regions angularly-separated from each other at a microphone array configured to transduce audio received from the search regions. The endpoint includes one or more cameras to capture video in a main field of view (FOV) that encompasses the main audio search region. The endpoint determines if audio originates from any of the main and side audio search regions based on the transduced audio and predetermined audio search criteria. If it is determined that audio originates from the side audio search region, the endpoint automatically switches from capturing video in the main FOV to one or more cameras to capture video in a side FOV that encompasses the side audio search region.