Set Up Turn-by-Turn Transcription

Permissions Required: Automation & AI > Launch Automation & AI Studio Script View, Create/Edit

Turn-by-Turn Transcription provides transcription of conversation audio. It sends the transcription in segments based on each turn in the conversation.

Add a Transcription Profile

A transcription profile defines the language you want to transcribe.

  1. In CXone, click the app selector and select Other > Automation & AI and select Transcription Profiles.

  2. Click Add Transcription Profile.

    The Add Transcription Profile page, where you can choose a transcription provider for a new profile.
  3. Enter a Transcription Profile Name. To differentiate this profile from your other profiles, use a name that highlights what makes this profile different. For example, you could name the profile for its intended use, the provider, or the language of the profile.
  4. Click the Transcription Profile Provider that you want to use.
  5. Click Next.
  6. Select the Language for this profile. Certain languages have an option to use Google's model for enhanced speech recognition. If the language you select has this option and you want to use it, click the Use Enhanced toggle that appears under the drop-down.
  7. Click Add.

Test a Transcription Profile

You can test a transcription profile to verify your selections. To test a profile, you must use a computer with a microphone.

  1. In CXone, click the app selector and select Other > Automation & AI and select Transcription Profiles.
  2. Select the transcription profile to test.

  3. Under Test, click the Microphone icon A microphone. and speak in the language you're testing. When you are done speaking, click the Stop icon A solid sqiare. . Turn-by-Turn Transcription automatically transcribes the words you spoke and displays the results under Test.

    The properties page for a transcription profile, where you can change the language and test the profile.
  4. Repeat the test as often as you need to. You can change languages if you want to. When you are finished, ensure that the Language selected is the one you want the profile to use.
  5. Click Save.

Configure the Studio Script to Use Transcription

Permissions Required:  Studio Scripts View, Create/Edit

If you're using Turn-by-Turn Transcription with a voice virtual agent, you don't need to follow these steps. The virtual agent integration process handles transcription data automatically. You don't need to include the Cloud Transcribe action in scripts for virtual agent integrations.

If you're using Turn-by-Turn Transcription to enable a text-only virtual agent to handle voice interactions, you must follow these steps. Your script must include the Cloud Transcribe action.

To use Turn-by-Turn Transcription for non-virtual agent applications, your script must have the Cloud Transcribe action. If you're transcribing multiple languages in the same script, you need to add one Cloud Transcribe action for each language. You also need a transcription profile for each language.

  1. In Studio, open the script you're working with.
  2. If the Cloud Transcribe action isn't already in the script, add it to the correct location.
  3. Assign a transcription profile to the action:
    1. Double-click Cloud Transcribe to open the Transcription Profiles page.
    2. Click the Selection icon next to the profile you want this action to use.
    3. Click Yes.
  4. Add a Snippet action before Cloud Transcribe. If you have more than one Cloud Transcribe, place it before the first one that appears in the script.
  5. Double-click Snippet and add the following code:
    DYNAMIC nextPromptBehaviors
<add behaviors as needed>
ASSIGN nextPromptBehaviorsJson="{nextPromptBehaviors.asjson()}"
  6. In the Snippet, replace <add behaviors as needed> with the behaviors you want Cloud Transcribe to use. You can learn about the behaviors or view a list of all available properties in alphabetical order.

  7. To configure Cloud Transcribe to pass on the returned transcription in JSON format to a voice-enabled text virtual agent: 

    1. Double-click Snippet to open the snippet editor.

    2. Add the following code to the snippet:

      DYNAMIC nextPromptSequence
ASSIGN nextPromptSequenceJSONout = "nextPromptSequence.asJSON()"

      The object defined here holds the transcription returned from the STT provider. Your script must be designed to pass this data to your voice-enabled text virtual agent.

    3. Right-click Cloud Transcribe and configure nextPromptSequence with the name of the variable that holds the returned transcript in JSON format. For example: transcriptVarNameJSON.

  8. Right-click Cloud Transcribe and modify the following properties: 

    • nextPromptBehavior: Configure with the name of the variable that holds the next prompt behaviors in JSON format. For example: nextPromptBehaviorsJson.

    • transcriptVarName (out): Configure with the name of the object that you defined in step 5. For example, transcriptVarName. This variable holds the returned transcript. Your script must be configured to handle the transcript.

  9. Save your script.