I have a customer that is dictating some rather unique and advanced call recording requirements.
All calls that get recorded must be sent to at least two NICE recorders simultaneously.
My understanding is that CUBE Media Proxy can help here. I have a PDI case open and it seems that is indeed accurate but had a couple follow-up questions and looking for validation:
It appears from the diagrams that CUBE Media Proxy accepts the RTP streams from the source and then relies on CUSP to handle the actual SIP signaling to the recording servers. Is that accurate? I'm assuming this is not an issue for a non-CCE environment?
It seems that CUBE Media Proxy requires a separate platform separate from the CUBE doing SIP Trunks. PDI mentioned a separate ISR or "Server".
What are the required specs for the ISR platform?
Are DSP resources required to process those media streams?
I'm not sure what they meant by server; but assuming that is maybe a virtual CUBE? If that is the case then I imagine DSPs are not required.
Customer wants has specific recording tone specifications that they are not willing to flex on. We have worked with a partner that handles custom CTI applications and it appears we have a solution using CUCM Silent Monitoring and Recording with BiB.
They want the tone to only be presented on calls with external parties (I think this may kill our current solution)
The tone has to be heard by both calling/called parties.
The tone must have an initial beep on call connection.
An additional tone 10-seconds later; with no subsequent beeps
The tone must have the following parameters:
Length of each tone: 1/5 of a second +/- 20%
Pitch of tone: 1400 cycles per second +/- 10%
Level of tone: Equal to average telephone talking signal strength
The are only to be enabled for a subset of recorded users.
My thoughts to accomplish this:
Utilize CUCM Silent and Monitoring Recording with Gateway Preferred Audio which utilizes the XMF CUBE API.
The question here is can the XMF recording handle multiple simultaneous recording destinations natively?
I'm assuming not since the recording destination seems to be based off of the recording profile's pattern and SIP trunk in CUCM. BUT - could you use a VIP of sort to front end multiple recording servers to trick it into handling it simultaneously?
If not, is it supported with CUBE Media Proxy? I do not see why not - it should not matter where the signaling/RTP is originated from so I'd think that this would work.
It seems that this would allow us to play the recording tones via the CUBE APIs. The question is...how is that configured and can we achieve the above requirements...?
This would then allow regular BiB recording to take place if no recording enabled gateway is in the call path thus not playing the tones.
To achieve the "play tone for only subset of users":
I imagine this will be some sort of custom API work here?
If we can associate this custom API is tied to a XMF I'm wondering if we could do two different XMF configurations tied to two different source loopback interfaces.