06-19-2017 02:36 AM
I'm new on this site and I only want to have your advices about a problem I have with GroupSeries 500 and Eagle Eye Director II. I had a session with a remote site and noticed that the cameras react to the voice of the remote participants. The consequence of that is that while the remote site is speaking and I'm just listening, the remote site doesn't see me anymore as the camera turns to my left screen (the one that shows the remote site). It's a 55'' screen with integrated speakers. Plus, I've noticed that the system keeps following even when I saying nothing. It keeps displaying me few seconds even when one of my collegue is speaking. Please help me to solve these problems.
Thank you for having read.
06-19-2017 04:42 AM
I thought a lot about the problem and came to think that using the fence technology could help if the system only react to sound emitted by the Polycom microphone. The problem is that EagleEye Director 2 also has his own audio input and I must recognize that I don't understand why since there's already a microphone. Do you know why?
06-19-2017 04:46 AM
Hello Bayonne Addo,
welcome to the Polycom Community.
It is always useful to include the currently used Software version as issues experienced or a question asked may already be addressed in a newer release.
This also allows yourself and others to check against current software release notes, Administrator Guides or FAQ post’s.
The above is also stated in the "Must Read First" and is the absolute minimum requirement every new post should include. .
In addition providing us with this basic information gives Polycom an idea what Software Versions are used in the field and avoids wasting time trying to troubleshoot issues which have already been addressed.
Please ensure to provide some feedback if this reply has helped you so other users can profit from your experience.
Polycom Global Services
06-19-2017 05:09 AM
Ok, Thank you SteffenBaierUK for taking your time to show me my mistake. I apologize for that and I will read the Must read first immediately. The info of the system I'm using are:
Group Series software version: 6.1.1-370109
Group Series hardware version: 18
EagleEye Director 2 software version: 184.108.40.206-13827
EagleEye Director 2 hardware version: A
Thank you for having read.
06-19-2017 09:40 AM
First, following the installation directions make absolutely certain that the audio between the RealPresence Group Series codec and the EED2 is connected.
Also make sure that audio level on the whatever you are using for audio output is at between 50 and 66% of the volume, Make other adjustments on the video system, not on the speakers or the TV or whatever else is being used.
06-20-2017 02:24 AM
Thank you Polycom Video for your answer. I'll assure that the output audio level is in the good gap. But I already can confirm that the audio between the two equipments is well connected. The EED2 reacts ALWAYS immediately in the direction of the speaking person but I measured sometimes near to 16s before the image appears on the screen. Sometimes it's 2s and some other times we reach 16s. We are in his gap: 2s, 4s,.. and sometimes 16s.
But please do you know why the EED2 also has integrated audio input since the system already come with a mic?
06-20-2017 06:10 AM - edited 06-20-2017 06:11 AM
The dynamic speaker locating technology at a high level works this way:
Using triangulation on the voice (thus the need for the microphone arrays on the EE Director) it finds the place within the room that the audio is coming from. It then uses face recognition technology to properly frame a person in the frame (the eyes at 1/3 of the way down the frame. Once it has been able to frame the person it will cut to that person. The audio input to the EED from the codec is to differentiate between near end and far end audio.
Some basic rules to optimize the EEDirector and the EE Producer are:
So if the person is not looking at the camera when speaking the system will have a much harder time to locate the person in the room. Also if they are too far away it will take more time to identify the person speaking. If the face is not lit so that the camera can find the facial features and frame the face it will take more time.
06-20-2017 07:17 AM
With the researches I've made and your precious feedback, I feel much better and confident to improve the system functionning.
But there's just only one thing I didn't succeed to explain. The EED2 was configured on frame speaker but even when I'm silent, the camera kept on showing me on the screen (even more than 5mn continuesly) and my colleague who was speaking had to wait sometimes more that 2s or 4s to be shown. I tried to move to another seat without saying a word and the system came to me again. The light and the fact that I'm a bit clearer than my colleague could be the cause?
06-21-2017 03:39 AM
To solve my problems I thought that using Fence technology could be great but I realize that the system offer only 2 parameters to do that: Enable acoustic fence and Acoustic fence sensitivity. Is it all? How can I configure the area t of the fence? I thought that it would be possible for example to chose a diameter around the mic. I'm in a room of 225 m². I tried different values even put 1 but the system react when we are far from it, on the corner of the room.