#any way to get chatgpt to output audio files using it's own voices i.e cove?

1 messages · Page 1 of 1 (latest)

viscid thunder
#

I was wondering if there was some trick to get chatgpt to say read an audio script using it's own set of voices and would save me some time and effort stitching something together on a third party site

jade wolfBOT
#

Hi @viscid thunder. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:

#

Currently, there is no API or programmatic method to generate audio files using ChatGPT’s in-app voices such as Cove, Ember, Juniper, etc. These voices are exclusive to the ChatGPT app’s Voice Mode and are not available through the OpenAI API or any other endpoint. The API provides a different set of built-in voices for text-to-speech (TTS), such as alloy, ash, ballad, coral, echo, fable, nova, onyx, sage, shimmer, and verse, but not the ChatGPT app voices like Cove or Juniper. You can preview and use these API voices via the Text to Speech API or OpenAI.fm.

To summarize:

  • ChatGPT app voices (Cove, Ember, etc.): Only available within the ChatGPT app’s Voice Mode, not accessible via API or for file export Voice Mode FAQ.
  • API voices (alloy, coral, shimmer, etc.): Available for programmatic TTS via the API, but these are a different set of voices Text to speech.

There is no “trick” or workaround documented in the knowledge sources to access or export the ChatGPT app’s voices for use in your own audio files. If you need to generate audio files programmatically, you’ll need to use the voices available through the API.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon: