Systems and methods for providing media content to supplement an exhibit from an image of an audience for the exhibit include an image capture device proximate the exhibit that captures an image of the audience that includes the faces of audience members. A facial recognition process is used to identify each audience member, and certain audience characteristics are determined from the captured image. Stored information about the identified audience members and/or the audience characteristic information are used to determine the media content to provide to a playback device for playback while the exhibit is viewed by the audience.