Speech Recognition

The Speech Recognition action allows you to convert an audio recording into text and return sentiment and tonality on any audio and text recorded. It could be used to provide transcripts of calls, measure the sentiment and tonality of an interaction, and more!

On the left the Speech Recognition action on a blank board, and on the right the Speech Recognition Inputs on the Configuration Panel

File Reference

This field allows you to select which channel you would like to analyze. Select ANI to record the caller, DNIS to record the receiver, and ICALL_#.CHANNEL_ID to record both channels. Depending on how you’ve built your flow, you may have additional channels to analyze, like a recording or an SMS.

 

Sentiment

When checked, sentiment analysis is performed on the audio file. This field is not editable and is the file name assigned to the recorded sentiment. For more information, check out the Sentiment page.

Note: There is a charge for each sentiment analysis performed.

 

Tonality

When checked, tonality analysis is performed on the audio file. This field is not editable and is the file name assigned to the recorded tonality. For more information, check out the Tonality page.

Note: There is a charge for each tonality analysis performed.

 

Transcribed Text

This field is not editable and is the file name assigned to the transcribed text.

 

Variables

The Speech Recognition action creates several variables. These variables can be used later on in a flow through the Send SMS, External Web Call, or Dial actions, to name a few. Simply copy and paste or drag and drop the file name into whatever field you would like to insert the Speech Recognition analysis.

Sample Speech Recognition variables that might appear in the Available Variables section of the Configuration Panel for downstream actions

Check out How to Use Variables in SmartFlows for more information about using variables in your flow.