A single action alone does not create a proper call scenario

Say enables you to convert the given text into a speech and play it in the currently active call.

The requests for all actions are the same as v1.1 except the removal of clientRequestId

A few things to note:

  • voiceProfile defines the characteristics of the generated speech. Choose one of the available Voice profiles to define the voice, gender and accent for the message you are sending. All available voice profiles can be found at here.
  • repetition controls how many times the speech is played back. The default value is 1.
  • speed controls the speed of the speech. (Accepted values are from 0.5 - 2, as a two-digit number. The default value is 1.
  • text The message body content to be converted to voice. Maximum voice message length is 3000 characters

Path for the new API

/POST{sub-account-id}/callflows - v1.2

Sample of a sayrequest

    "action": "say",
    "params": {
        "text": "Hello, this is a Voice Message."
        "voiceProfile": "en-US-ZiraRUS"
        "repetition": 1,
        "speed": 1