As an Amazon Associate I earn from qualifying purchases.

Cisco UC Call Recording with MediaSense

Cisco's MediaSense, introduced in 2011, is their solution for multimedia capture, streaming and recording.

This article provides a brief overview of the solution, describes the MediaSense deployment models and gives information on how to improve upon the free user interface that Cisco provides with MediaSense at no additional cost.  Cisco also provides extensive design, installation and upgrade, configuration, maintenance and troubleshooting resources online.  While this article is focusing specifically on call recording and the "Search and Play" interface, all characteristics and features of MediaSense can be explored in the design guide.

Let's start with some call recording history first…

History of Cisco UC Call Recording


Before 2011, the only way to record phone calls in the Cisco IP telephony environment was to use one of several 3rd party products. Most of these solutions used “SPAN-based” recording.  When SPAN (Switched Port Analyzer) is enabled on a switch port (or VLAN), the switch copies all network packets to another port where a recording solution detects the VoIP packets stores them as audio files.

Some also supported a more efficient "audio forking" approach provided by Cisco through the use of bridge functionality provided within the IP phone itself.  Most Cisco IP phones contain the embedded conference bridge (“built-in bridge or BiB”), which, with proper CUCM configuration, will fork the audio streams of the phone call (for example, one for the agent voice and one for the customer voice) to the recording server.

Built-in-Bridge (BiB) and SPAN Recording


Introduction of Cisco MediaSense


Near the end of 2010, Cisco entered the call recording market and performed 3 great steps:

  1. Cisco introduced the MediaSense product that captures audio-streams duplicated by Cisco IP phone built-in bridge;
  2. At the same time Cisco released the version 15.2 of IOS (software that runs on Cisco routers) with media forking feature that duplicates the phone call RTP streams to Cisco MediaSense for recording;
  3. Cisco positioned MediaSense as the recording platform with API to 3rd party apps.  MediaSense works on the network layer capturing and storing the phone calls but you need to turn to 3rd party vendors for business applications that work with call recordings, like speech analytics, quality management, agent training, etc.
Cisco MediaSense now supports the capture of media streams forked via:

  • Cisco IP phones (models with built-in bridge) – to record employees’ Cisco phones;
  • Cisco Unified Border Element (application run on Cisco IOS) – to record 3rd party SIP devices and external-to-external phone calls.
Cisco MediaSense Recording


Unfortunately, from the users’ perspective, MediaSense only provides a simple “Search and Play” web-interface that is not as feature-rich as most desire:

Cisco MediaSense Search-n-Play User Interface


The "Search and Play" interface is a free web-application provided by Cisco (with the source code) as an example of how to use the Cisco MediaSense API – this password-protected web-page allows user to perform basic search and play call recordings.

Cisco MediaSense API Enables Innovation


As a result of the API that Cisco makes available, most call recording software vendors now support integration with Cisco MediaSense to provide clients with more convenient user interfaces to access the call recording archive with features like:

  • search recordings by date, time, phone number, employee’s name and client name;
  • indication of a call direction (incoming/outgoing);
  • logical grouping of records in case of call routing, putting on hold or parking and organizing a conference with several participants;
  • email notification about new recorded call;
  • search and playback interface on Cisco IP phone;         
  • option to playback recordings during a phone call (all call participants will hear the recording).

Free Call Recording UI For Cisco MediaSense


To provide additional features beyond what Cisco provides "out of the box" with MediaSense, one Cisco Solution Partner offers a free interface to Cisco MediaSense that contains all of the features listed above.  

Additionally, their commercial “Aurus PhoneUP” application suite contains both a Cisco MediaSense interface and a complete call recording solution for Cisco UCM and BE 6000/7000.

7 comments:

  1. I have not had very good luck with CUBE forking. MediaSense seems pretty rock solid, and the forking seems to work well too, the problem is the plethora of bugs within the CUBE code. If you run a large portion of your enterprise calls through a CUBE, and the CUBE (or cube cluster) fails due to one of the umpteen billion bugs, it is devastating to the call center. BIB is nice, but if you have people working from home with a VPN, multiple streams from the remote phone has their own issues when traversing the internet. I'm actually looking at going with a span analyzer and ditching MS and CUBE forking.

    ReplyDelete
  2. I don't see a section on there for tools to use, so I'm going to list a few that I would not be able to work without. You would be surprised at how many people have never even heard of TranslatorX. This is invaluable for troubleshooting call setup. Notepad++ is indispensable not only for its tabbed text editing and ability to highlight different terms in different colors, but its ability to search all files in a directory for a text string makes it an amazing tool for finding stuff inside RTMT trace logs. Another great tool for trace analysis is Agent Ransack. I use this one less that I used to, but still find it useful as a companion to Notepadd++ file search feature. Next up is WinMerge which allows you to compare configurations side by side and easily see the differences in the configs. And finally, if you haven't become intimate with Wireshark yet, you need to.

    ReplyDelete
  3. Great comment and content! This would be a great topic for a future post. I appreciate you taking the time to share this!

    ReplyDelete
  4. Heresy! Bugs in IOS? :)

    The real world experience here is appreciated - buyer beware, validate in the lab and take your vitamins. Voice through VPN certainly isn't with it's added complications - sprinkle in virtualized desktops as well and it's a recipe for a headache.

    ReplyDelete
  5. Hi,
    Nice post - You mention External to External. Is it possible to record this call scenario: PSTN>TDM>CUBE>CUCM>IPPhone>Call Forward>CUBE>Mobile Phone. External call to internal extension which is forwarded to Mobile phone. Sim Ring.

    ReplyDelete
  6. It seems as though you should be able to record the streams using media forking on the CUBE as long as you meet the Cisco requirements, IOS, config, etc. Because of the call path, I think you may end up with two legs for the inbound call from PSTN and then two additional legs for the forwarded call to a Mobile phone. The result could be two nearly duplicate media files on MediaSense. It's an interesting scenario and without testing, nothing is certain. I do think you'll get the streams recorded though, perhaps duplicated.

    Decent config guide: Cisco MediaSense and Cisco Unified Border Element (CUBE) for Network based Media Recording Solution https://shar.es/1qeYe3

    ReplyDelete
  7. Actually, as long as your IP phone has the BiB, I would use that instead. It's more efficient and the CUBE media forking can be fickle as Nick mentioned above. External-to-external refers to calls that are routed by the CUBE and do not reach CUCM/IP phone w BiB, I believe.

    ReplyDelete