cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Announcements

Community Helping Community

2152
Views
4
Helpful
10
Replies

Starting with MediaSense Call Recording

Hi Everyone,

Hope you will be fine and enjoying your life. I recently started working in MedisSense API. I have gone through its documentation and Now I am actually turning to its Implementation. I want to record a call made to particular extension. For this I used this API call:

http://---.---.---.---:8440/ora/controlService/control/startRecording

And Body:

Header: Content-Type: application/json & mentioned JSESSIONID too.

Body:

{"requestParameters": {

"deviceRef": "42021",

"mediaStreams": [

{

"mediaType": "AUDIO"

}

]

}

After making this API call, I make call to extension mentioned in body of this request that is '42021'.

I have tried this using poster. I didn't get any response. I want to know that this is how we record a dialogue ? where the recorded file is saved ? I need sample Application from which I may take help. The one uploaded by cisco is not sufficient for me. Ant tutorial, sample Application and help will be highly appreciated.


Thanks,

1 ACCEPTED SOLUTION

Accepted Solutions
Highlighted
Cisco Employee

Re: Starting with MediaSense Call Recording

This is a common misunderstanding.  The startRecord API is unfortunately named; it's probably NOT what you want to use.  In fact it is rarely used by anyone.

There are two kinds of Direct recording in MediaSense: Direct Inbound and Direct Outbound.  These are described in the SRND and the User's Guide.  With Direct Inbound, you configure a trunk from CUCM to MediaSense and assign it a dial pattern.  Then from any CUCM phone, you dial that pattern.  MediaSense answers, and simply records whatever you say.  You will be able to see the recording in the Search and Play application, but it is NOT the recording of a 2-way conversation.

Direct Outbound is exactly the same thing, except that MediaSense calls the phone rather than the phone calling MediaSense.  They way you get MediaSense to call the phone is by issuing the startRecord API.  Again, this does NOT record a 2-way conversation; it only records what the phone user speaks into the phone.  (And by the way, if the phone supports video, MediaSense will record the video as well.)

For your initial testing, you might want to try the Direct Inbound method described above.   But then you'll want to quickly move on to configuring CUCM recording profiles and other structures, and configure your phone to "always record".  (Make sure you are using a phone that supports media forking!).  Then you'll start capturing 2-way conversations.

View solution in original post

10 REPLIES 10
Highlighted
Cisco Employee

Re: Starting with MediaSense Call Recording

This is a common misunderstanding.  The startRecord API is unfortunately named; it's probably NOT what you want to use.  In fact it is rarely used by anyone.

There are two kinds of Direct recording in MediaSense: Direct Inbound and Direct Outbound.  These are described in the SRND and the User's Guide.  With Direct Inbound, you configure a trunk from CUCM to MediaSense and assign it a dial pattern.  Then from any CUCM phone, you dial that pattern.  MediaSense answers, and simply records whatever you say.  You will be able to see the recording in the Search and Play application, but it is NOT the recording of a 2-way conversation.

Direct Outbound is exactly the same thing, except that MediaSense calls the phone rather than the phone calling MediaSense.  They way you get MediaSense to call the phone is by issuing the startRecord API.  Again, this does NOT record a 2-way conversation; it only records what the phone user speaks into the phone.  (And by the way, if the phone supports video, MediaSense will record the video as well.)

For your initial testing, you might want to try the Direct Inbound method described above.   But then you'll want to quickly move on to configuring CUCM recording profiles and other structures, and configure your phone to "always record".  (Make sure you are using a phone that supports media forking!).  Then you'll start capturing 2-way conversations.

View solution in original post

Cisco Employee

Re: Starting with MediaSense Call Recording

One more thing - the non-response from Poster.  That sounds like you did not take steps to accept the MediaSense certificate in your browser first.  This is described in the MediaSense Developer's Guide, under "Building your Application."

Re: Starting with MediaSense Call Recording

Yes, I added the certificate in java for using this API Programatically but didn't add certificate in browser but now I have added and I can make HTTPS requests from poster too.

Re: Starting with MediaSense Call Recording

Then I have the following questions:

1. StartRecording API is used just for outbound calls recording ? If yes then it means that outbound call recordings can be done by running this API call withour any configurations on MediaSense server as I have APIUser and make calls by this.

2. Will we have two recordings for each call, one for inbound and other for outbound ? Is this not possible to have single recording against each call containing both what phone user and agent speaks ?

Thanks for your assistance in making me things clear :-)

Frequent Contributor

Re: Starting with MediaSense Call Recording

StartRecordingAPI is used for blog recordings. Per the MediaSense Developer Guide section on Mapping a Session to a Recording:

Several recordings are possible using the Cisco MediaSense platform:

Forked media from a Cisco IP Phone or Cisco Unified Border Element (CUBE) device. This recording has two audio channels.

Direct call to/from the Cisco MediaSense platform to any phone. This recording has one audio channel and one optional video

channel. These recordings are referred to as Blog recordings in this document

Forked media calls are retrieved using the Session API. Per the section Mapping a Media Stream to a Track:

A media stream refers to the packets going through an audio channel or video channel in a live or recorded session.

A track identifies each media stream and quantifies it with additional data such as participants, duration, startDate, and a trackNumber. Track

0 contains media streamed from the forking device, and Track 1 contains media streamed to the forking device. In a

sessionEvent , each track is associated with one participant.

Re: Starting with MediaSense Call Recording

dlender jwolf

I want to configure my MediaSense server for calls recordings. What steps I need to follow ? Or is there any document that may guide me regarding configuring my MediaSense server for call recordings ?

Thanks,

Frequent Contributor

Re: Starting with MediaSense Call Recording

I am not aware of any guide documentation on how to configure Callmanager and MediaSense to record calls.

If you are a Cisco Partner you may wish to utilize the PDI Helpdesk to assist you in this configuration.

http://www.cisco.com/web/partners/tools/pdihd.html

Frequent Contributor

Re: Starting with MediaSense Call Recording

See the MediaSense User Guide here

http://www.cisco.com/c/en/us/td/docs/voice_ip_comm/cust_contact/contact_center/mediasense/911/user_guide/CUMS_BK_M5B01864_00_ms-user-guide-911/CUMS_BK_M5B01864_00_ms-user-guide-911_chapter_011.html#CUMS_CN_I1F747F8_00

See Administer and Configure MediaSense.

Frequent Contributor

Re: Starting with MediaSense Call Recording

Also your body is missing a '}'

http://-.-.-.-:8440/ora/controlService/control/startRecording<http://192.168.200.204:8440/ora/controlService/control/startRecording>

And Body:

Header: Content-Type: application/json & mentioned JSESSIONID too.

Body:

{"requestParameters": {

"deviceRef": "42021",

"mediaStreams": [

{

"mediaType": "AUDIO"

}

]

}

}

Re: Starting with MediaSense Call Recording

yes, In my actual code it's all right. I didn't copy completely code here and missed a closing '}' brace.

CreatePlease to create content
Content for Community-Ad
FusionCharts will render here