8831 Conference Phone - Voice Echo

[demerson111]

Hello all,

Let me preface this problem by stating I am fairly new to UCS (6 months), so if anyone offers solutions to this issue, please use laymen's terms. I'm self-taught, but have a very general idea of how things work in this world.

I have a 8831 conference phone which is experiencing a feedback voice echo on the distant end during the first 10-15 seconds of a call. After the 10 to 15 seconds, the issue is corrected and call quality is otherwise good. This occurs when I use both wired and wireless mics with the base station. I was pretty much resigned to the fact that this is just something that we have to live with, because honestly it is a very short time, and the phone otherwise works perfectly. However, we don't experience this problem on the older model 7937.

Several months back, we suffered an outage which resulted in our network team restoring CUCM from a previous configuration in DRS. This was all done prior to me even being involved with call manager, as eventually the previous administrator got fired, and I just kind of inherited this mess. My point here is that my boss insists that prior to the recovery, this 8831 used the wireless mics and the call quality was perfect and did not experience this 10-15 second initial echo adjustment.

I have already upgraded the firmware to sip8831.10-3-1SR2-2, as recommended by Cisco, and replaced it with a new base station. The initial voice echo on the other end is still present. At this point, I am out of ideas to resolve this issue, so I welcome any suggestions.

We're using CUCM 9.1.2. If anyone needs more information for an assessment, please advise. Thank you!

[Jaime Valencia]

A DRS backup/restore has nothing to do with voice issues like echo.

Have you verified the room where the endpoints are is not the one causing the echo??

[demerson111]

I would agree with you on the DRS, but am using that as it is my only point of reference given to me by the previous admin. After restoration of an old configuration, this is when the problem started - leading me to believe that there is some parameter in call manager that maybe I am overlooking.

That aside, I have called numerous phones, from different rooms, all with the same result. On the far end, when the caller starts to speak, the mics on my end are picking up that person's voice and causing a feedback loop resulting in that person hearing an echo of their own voice. After 10 seconds, the phone corrects the problem.

In my opinion, waiting 10 seconds is almost trivial, but my boss won't accept this answer. These phones are expensive, and we don't experience this problem with the 7937 conference phone using SCCP. I'm throwing this issue out here to see if anyone else has had this problem with the 8831.

Same room, same numbers dialed from a 7937 = no initial echo. Any ideas?

[Manish Gogna]

You may check the following doc for voice quality troubleshooting basics

https://supportforums.cisco.com/document/101961/how-troubleshoot-voice-quality-issues-ucm-environment-bad-sound-no-audio

If the issue persists i recommend opening a TAC case since it will require identifying the call flow, topology in detail and colecting debugs / pcaps and analyzing them.

Manish

[demerson111]

Thank you Manish, this document is very helpful for general troubleshooting. A previous TAC case was opened for this problem by my predecessor, and the engineer recommended a firmware upgrade. After the firmware upgrade, I made a single test call and everything appeared to be working fine, so I went ahead and closed it assuming it would be good.

However, the problem re-emerged and seems to be intermittent at this point. It is a subtle problem that I would probably live with if my supervisor wasn't on my back about it. Here are the differences as I see them:

-With a conference room phone 7937 using SCCP, the call quality seems noticeably better, but when comparing it with the 8831 SIP phone, both have an avg MOS LQK of 4.5, and negligible packet loss, but for some reason the 7937 sounds better

-As I mentioned earlier, the catalyst to this problem was restoring from an old backup from DRS. I am just speculating - it sounds counterproductive - but I am wondering if using an older version firmware would fix it?

It's a very strange problem, but I think I will try an older phone load to see if that resolves the problem. Can these phones use SCCP?

I'll post back my results, unless anyone else has another suggestion.