TTS Voice Customization

• Voice: Name of the voice to use for synthesis. The default is Charlie. Other available voices include Rebecca, Prospector, Vampire, and Cooper. Select Fetch Voices on the WitTtsSpeaker actor to refresh the list of available voices from Wit.ai⁠.
• Style: Style of speaking, such as soft or formal. The same styles are not available for every voice. Select Fetch Voices to see the styles available for the selected voice.
• Speed: How fast the text is spoken, as a percentage of the voice speed as originally recorded. Values range from 50% to 200%, with 100% as the default.
• Pitch: The pitch of the voice audio, as a percentage of the original voice. Values range from 25% to 400%, with 100% as the default.
• Gain: The audio gain, in percentages from 0% to 100%, with 50% as the default.
  

• Text-to-Speech Overview: Setup and configuration for Voice SDK text-to-speech
• TTS Cache Options: Runtime and disk cache settings for TTS clips