An audio scene is captured in multiple audio channels by distributed microphones. Based on at least one of a viewpoint of a user and at least one preference of a user, two or more of those multiple audio channels are selected from among those multiple audio channels; those selected two or more audio channels are prioritized based on at least one of preferences of the user, requirements of a software application of a device operated by the user, and requirements of operating conditions of the device. After that prioritizing, at least one of the selected two or more audio channels is encoded into a coded monitor mix; which is sent to the device over a transmission link.