Documentation

Looking for something in particular?

Automation V3: Utterance Generator

This API enables users to automatically generate a set of sample utterances for a specified intent. Below are the key details and behaviors:

Inputs

  • Intent (Text, Mandatory)
    The intent must be clearly defined, appropriate, and actionable. Providing well-phrased intent text helps ensure accurate utterance generation.
  • Count (Integer, Mandatory)
    • Specifies the approximate number of utterances to generate.
    • Acceptable range: 1–100.
    • Defaults to 10 if not provided.

Output

  • The generated utterances are saved as a CSV file in the templates folder.

File naming convention:
Utterance_Data_<8DigitUniqueId>_<Timestamp>.csv

The CSV always includes a header row with:
Utterances

Each subsequent row contains a unique utterance generated for the specified intent.

Result Handling

  • The file name is stored in a test case variable, using the “Result Name” parameter provided by the user. This allows further automation  within Provar.

Failure Scenarios

If the AI is unable to generate utterances:

For example, due to an unclear or inappropriate intent—the output CSV will still be created with the same structure but with the content below:
Sorry, I can only generate utterances for clear, appropriate, and  actionable intent topics. Please rephrase your request.

Usage Notes

  • To improve success rates, users should rephrase vague or overly broad intents to be more specific.
  • The output format is designed to be easily consumed in Provar tests without additional transformation.

Intent Validator

This API validates whether a given utterance accurately expresses a specified intent. It provides a consistent way to test Natural Language Understanding behavior and confirm intent recognition.

Inputs

  • Intent (Text, Mandatory)
    The intent that the utterance will be evaluated against.
  • Utterance (Text, Mandatory)
    The user input or example phrase to validate.

Output

The API returns a single result value indicating whether the utterance matches the intent:

  • true
    The utterance clearly expresses the intent.
  • false
    The utterance clearly does not express the intent.
  • unknown
    The relationship between the utterance and the intent is:
    • Unclear or ambiguous
    • Lacking sufficient context
    • Inappropriate, nonsensical, irrelevant, or out of scope

Important:

  • The response will be only one of these values: true, false, or unknown.
  • No explanations or justifications should be included in the response.

Examples

IntentUtteranceOutput
book flightI need to book a flight to Mumbai tonight.true
cancel reservationCan I make a new reservation for next week?false
track orderWhere is my package? I ordered it yesterday.true
make paymentShould I bring cash or card?unknown
reset passwordYou’re so stupid, fix my life!unknown
find nearby restaurantsAliens are watching me through the microwave.unknown
get weather updateWhat’s the best way to cook pasta?unknown

Result Handling

  • The result value is saved into the Result Name attribute of the TestAPI.
  • This allows further automated processing or assertions in the test workflow.


Feedback

Was this article helpful for you?

Trying to raise a case with our support team?

We use cookies to better understand how our website is used so we can tailor content for you. For more information about the different cookies we use please take a look at our Privacy Policy.

Scroll to Top