Answer

Speaker Spotlight can operate as either audio-only or vision-assisted. The difference is in whether an audience-facing conductor camera is used to support the ceiling microphones, allowing VisionSuite to provide a vision layer on top of the audio coordinate data that fuses participant locations and provides improved accuracy.

Audio-Only Speaker Spotlight

Without visual confirmation, shots are composed entirely from microphone coordinates, which may not precisely reflect the speaker’s actual position.

• When a speaker stops talking, their location is eventually forgotten by VisionSuite
• If they speak again, a new shot is created and a different camera may be used
• This results in more variation in framing between speaking turns

Vision-Assisted Speaker Spotlight

An audience-facing conductor camera adds visual room awareness to aid every shot decision, matching audio coordinates with participant detections and anchoring framing to the active speaker’s actual position in the room.

• Visual detections give VisionSuite persistent locations of participants, even during silence
• This produces more centered framing, fewer unnecessary camera switches, and smoother transitions between speakers between speaking turns

The audience-facing conductor camera, however, is constrained by a depth limit and its field of view. For more information, see FAQ | What are the maximum camera distances for VisionSuite Native?

VisionSuite supports up to 25 audience member detections on the conductor camera’s vision stream and up to 5 presenter detections on the presenter tracking camera's stream, for a maximum of 30 total participant detections. Participants ignored by Exclusion Zones or a Tracking Zone do not count as part of the participant limit, meaning that back rows can be excluded and served by audio-only Speaker Spotlight.