Changing the speech provider

At any stage of the call, the bot can dynamically change the speech provider (speech-to-text or text-to-speech) of the call dynamically. The provider change can be done for the entire call duration, or for the current activity (the current text/audio that is played by the bot).

For example, some speech-to-text providers may be better at recognizing dates while others are better at recognizing streets. The bot can change the active speech-to-text or text-to-speech during the call.

---

How do I use it?

At any stage of the call, the bot can dynamically change the speech provider (speech-to-text or text-to-speech) of the call dynamically. The provider change can be done for part or an entire call duration (activity/session params). The following is an example of dynamically changing both the text-to-speech and speech-to-text providers using the ttsProvider and sttProvider activityParams.

This feature is configured on VoiceAI Connect by the Administrator, using the additionalSttProviders and additionalTtsProviders bot parameters (see Configuring multiple speech providers).

Example:

The following shows an example of a bot configuration for changing text-to-speech and speech-to-text providers:

{ 
  "type": "message", 
  "text": "Please tell me your date of birth", 
  "activityParams": { 
    "ttsProvider": "my_aws_provider", 
    "sttProvider": "my_google_provider" 
  } 
}