Skip to main content
POST
Clones a voice from a short audio sample via Cartesia and stores it as an organization-scoped custom voice. The returned voice can then be assigned to any agent in the organization like a built-in Cartesia voice. The request is multipart/form-data (it carries an audio file).

​
Form Fields

​
file
required
The audio sample to clone. Supported types: mp3, wav, flac, ogg, webm. Maximum size 4 MB. A 10–30 second clip works best.
​
string
required
A name for the voice (unique within the organization).
​
string
default:"en"
Clone language. One of Cartesia’s supported codes: en, fr, de, es, pt, zh, ja, hi, it, ko, nl, pl, ru, sv, tr, tl, bg, ro, ar, cs, el, fi, hr, ms, sk, da, ta, uk, hu, no, vi, bn, th, he, ka, id, te, gu, kn, ml, mr, pa. Stored with the voice and applied as the agent’s voice_settings.language.
​
string
required
Must be the string "true" to attest you have the right and consent to clone this voice.
​
string
required
The organization the voice belongs to.
​
string
Optional description.

​
Response

Returns the created voice as a voice option (provider: "cartesia", custom: true). Its id is the Cartesia voice id to store in an agent’s voiceSettings.voice_id.

​
Errors