Was wondering if it is possible to connect a TTS provider thats not currently out of the box supported by VAPI. Looking at Minimaxi as I experimented and thought it was pretty good for my use case.

Currently, Vapi does not provide a documented way to directly integrate a custom TTS provider (like Minimaxi) if it is not already supported out of the box. The platform supports bringing your own models for transcription and LLMs, but there is no mention of a similar mechanism for TTS. If you need this feature, I recommend reaching out to [Vapi support](https://docs.vapi.ai/support) to discuss your use case or request this as a feature. Sources: - [Provider Keys – Vapi Documentation](https://docs.vapi.ai/customization/provider-keys) - [Custom Transcriber – Vapi Documentation](https://docs.vapi.ai/customization/custom-transcriber) - [FAQ – Vapi Documentation](https://docs.vapi.ai/faq)

Would like to request this as a feature

Yes, you can integrate a custom TTS provider like Minimaxi with Vapi by setting up your own TTS server and configuring your assistant to use it as a custom voice provider.

oh its this? https://docs.vapi.ai/customization/provider-keys

Look for custom voice at you will get the details. https://cdn.discordapp.com/attachments/1374305861865967627/1374715769014976632/image.png?ex=682f0f54&is=682dbdd4&hm=00790fa73f97c4b9a67ee31ea04bae9abe859ffba9820f79ed4a9791201276b4&

oh alright thanks!

Glad the issue was resolved. Please reach out to us if you have any other questions regarding this issue

Hey sorry so when I make the API call, still not sure where to pass the body? shows the request example bht where do i place the body?

Your request will look something like this:

curl -X POST https://api.vapi.ai/assistant \
     -H "Authorization: Bearer " \
     -H "Content-Type: application/json" \
     -d '{
  "voice": {
    "provider": "custom-voice",
    "server": {}
  }
}'

You can also review this reference guide and build a template to gain a better understanding of the request

as in the body for the server under voice? ie. the servier is an api and i need pass a body

Hey Timmy, you can check out this guide on how to set up custom text-to-speech with VAPI, and let me know your feedback on this document. After Your feedback, I'll deploy this document onto VAPI's official docs.

Thanks will try it out