Avaya CU360 Release Notes page 23

Collaboration unit
Hide thumbs Also See for CU360:
Table of Contents

Advertisement

© 2021 Avaya Inc. All rights reserved
Track Speaker
Thanks to the combination of audio tracking and face recognition, based on AI methods, the
CU360 can identify the active speaker position in a typical huddle room with a max of four seated
persons in front of the embedded camera (in a field of view of around 120°), and automatically
digitally pan-tilt-zoom the camera to provide the best framing for the speaker.
Once CU360 is placed on top of monitor, please check the lens alignment of the CU360
embedded HD camera to the device front panel. To achieve satisfactory results for track speaker,
only tilt angle (up or down) can be manually adjusted. Do not adjust the pan angle (left or right
direction).
When no audio or face is detected for around 20 seconds, the camera will automatically zoom to a
default position, corresponding to about 1.5x zoom factor.
If a face is detected for several times in the same position, the above timeout will be increase to 2
minutes in places of 20 seconds, and the tracking will remain on the last face detected, if still in
the same position.
Figure 4:
Speaker Tracking Disabled vs Enabled
Release Notes for Avaya Collaboration Unit CU360
Important
When the mic is muted or the camera is in privacy, the Speaker Tracking feature will stop
(only for the CU360 app). In this case, after a few seconds, the camera will go back to a
central wide-angle position. If the same person is speaking for several times in a row, the
interval to go back to a central position will increase to a couple of minutes, and the
central position will be restored only if the last detected face has changed position.
If in the room there are sources of noises, sometimes the camera is digitally pan/tilted to
the wrong direction or the algorithm needs more time to detect the speaker. It is also
possible that the audio algorithm is able to detect the speaker, but the face is not
recognized (too far or not in front position). In those cases:
Check that the camera is not manually panned left or right respect to the
o
central position.
Check that there are no sources of noise in the room.
o
Check that the room is properly lighted.
o
Assure that the distance of the talkers from the camera is not more that 3.0-
o
3.5 meters, and that the face of the talker is in front of the camera.
Keep talking for a few seconds with your face towards the camera to have the
o
camera detect you.
See Optimal Huddle Room Setup for supported arrangement of the room. Up to 4
different faces can be recognized for Speaker Tracking.
The audio detection algorithm for Speaker Tracking is now available also when a BT or
USB microphone is used.
Avaya CU360 Setup and Control 23

Advertisement

Table of Contents
loading

Table of Contents