With the previously used software, we had an API that provided the latest recognitions in a layout designed for a TV screen. It included customizable options to define how many recognitions to display per slide, how frequently slides should change, and how many of the latest recognitions to show.

For reference, here is a link that illustrates this functionality: