Add text-to-voice API references

[louisjoecodes]

• Introduce text-to-voice routes in the api documentation [Blocked by Fern bug] TODO:
• [x] Make TTV doc improvement suggestions
• [ ] Get feedback on proposed description/summary changes
• [ ] Update back-end with new descriptions
• [ ] Run the workflow to automatically update the OpenAPI specification once live

• [ ] Rebase this PR

Route: /v1/text-to-voice/create-previews

Description: Generate a voice preview from a description

Summary: Generates custom voice previews based on a provided voice description. The response includes a list of voice previews, each containing a generated_voice_id and a sample of the voice as a base64-encoded MP3 audio file. If you find a preview you like and want to create a permanent voice, use /v1/text-to-voice/create-voice-from-preview with the corresponding generated_voice_id.

---

Route: /v1/text-to-voice/create-voice-from-preview

Description: Create a new voice from a voice preview

Summary: Creates a new voice using a previously generated voice preview. Use this endpoint after obtaining a generated_voice_id from /v1/text-to-voice/create-previews.

---

[louisjoecodes]

Hey @samsklar11 & @lharries,

The API routes for TTV will be made public soon, but in the meantime, I’m proposing some updates above to the documentation for these routes.

Here’s what’s currently defined on the backend: https://elevenlabs-docs-ed-add-ttv-docs.mintlify.app/api-reference/ttv-create-previews

You can find my proposed summary/description for the two routes above. Once you’ve reviewed them, I’ll request the team to update the descriptions at the API level.

Thanks!

[lacop11]

FYI in the meantime I've merged #451, mostly based on your suggestions. There is a pending change to the backend schema, we can regenerate once released. And the generated snippets are not using SDK, maybe they didn't pick up the change yet and we need to run it later.