Speech Analytics

Overview

Speech recognition and natural language processing technologies improved rapidly in terms of accuracy and cost efficiency. Now it is more practical than ever to enhance your telephony services with speech analytics. Voxbone’s speech analytics technology partners provide high quality speech recognition and analytics services for a wide variety of applications like transcription, monitoring for sales effectiveness or dealing with sensitive data.

With readily available integrations built right into the network, Voxbone’s speech recognition optimised cloud recorder lets you easily add speech analytics to Voxbone’s trusted DID services.

For users looking for the easiest way to record, they can use a Voxbone DID. But we recognise Voxbone may be one of many providers to our customers. In our mission to make speech analytics strangely simple but also provide 100% call coverage even if Voxbone is not handling the call. To that end, we introduced SIPREC support last year so that we can handle any call, even if it does not normally traverse Voxbone. Check our blog article to learn more about how it works.

 

Service Activation

In order to see speech analytics features and configuration options, you first need to activate the feature under your account on our Customer Portal

To activate the service, log in to our customer portal and click “Request Activation” for Speech Analytics feature on your service overview. Once the activation request is submitted, you will receive a notification when the service is enabled on your account.

 

Sending Media to Voxbone

Speech Analytics uses the Voxbone Labs Virtual POPs. If your network is configured to limit incoming traffic from defined IP addresses then you must add the following addresses to your whitelist:

SIP Signaling Addresses:
146.148.10.67
35.194.231.175

Depending on your network configuration, it may additionally be necessary to explicitly whitelist the IP addresses that media (RTP traffic) will be coming from:

RTP Media Addresses:
35.195.209.181
35.195.214.30
35.194.198.160
35.201.231.65

Using SIPREC to Send Media

The idea behind SIPREC is that users should be able record calls without interfering with the original call. In most situations, metadata about the caller and callee are needed to identify customers and employees associated with the call. So we’re talking about the call center agent name and the caller ID.

The SIPREC architecture achieves these goals with three main components:

  1. Session Recording Client (SRC): This is a network element that copies the calls and metadata and sends them to be recorded.
  2. Session Recording Server (SRS): The recorder that receives the calls will save them to disk, and ties in any associated metadata.
  3. Metadata: SIPREC standardizes common fields that can be sent in a standard XML format

For users looking for the easiest way to record, they can use a Voxbone DID. But we recognize Voxbone may be one of many providers to our customers. In our mission to make speech analytics strangely simple but also provide 100% call coverage even if Voxbone is not handling the call. To that end, we introduced SIPREC support last year so that we can handle any call, even if it does not normally traverse Voxbone

If you’d like to use SIPREC to send media to us for speech analytics, Contact Us to get started.

Enable Analytics on Voxbone DIDs

After you make sure you’ve tested sending media to our analytics platform, you can start recording and speech analytics on a Voxbone DID. This can be done by creating a new Voice URI with Speech Analytics enabled. You can also link a Speech Analytics URI with a standard Voice URI as backup.

Enabling recording and speech analytics on a Voxbone DID will start sending media to our analytics platform immediately. The call will then be forwarded to your VoIP platform with the originally configured Voice URI destination.

Step 1 : Go to Configure > Configure Voice URIs on our customer portal

Step 2 : Click “+ New” to create a new URI and enable Speech Analytics

Step 3 : In the popup window, check the “Speech Analytics” feature

Step 4 : Pick the language and enter the Voice URI you want to enable speech analytics on


Step 5 : Link the newly created analytics URI to one of your DIDs. You can search for analytics URIs by looking for domain “sip.voxrec.io” in the Voice URI

Step 6 : Apply configuration! As soon as the configurations are applied, you will start seeing new calls coming in on the “Account > Speech Analytics History” on navigation menu

Review and Export Results

 

Once Speech Analytics is enabled on a Voxbone DID, or you started sending media via SIPREC, you will be able to review them on our Speech Analytics History page. On this page, you can;

  • Review basic CDR data of calls with speech analytics
  • Listen to and/or download recorded files of a caller-only, callee-only or the mixed stereo channel file
  • Review speech-to-text output of the call in conversational format
  • See sentiment score per sentence and sentiment trend throughout the call
  • See talk time analysis and overtalk incidents where a participant interrupted the other
  • Perform keyword or phrase search and filter calls by advanced speech analytics data such as sentiment score or talk time
  • Export all or filtered speech analytics data into a CSV file to use with any Business Intelligence tool to easily build dashboards such as Tableau, Microsoft Power BI, etc.
  • Download JSON response from our API to get started on working with speech analytics data in your own apps.
  • Export an equivalent cURL or relevant API request body

Integrate Analytics data into your platforms

You can also perform search, filtering and data export via using our analytics API and build your own dashboards, or integrate speech analytics data into your own platforms or business processes.

Check out our API documentation to see example requests, available endpoints and sample responses.