API Reference

Voice Cloning

Beta

Create a personalised voice from a short audio sample. Returns a voice_id you can pass to the TTS endpoint to synthesize speech in the cloned voice.

Endpoint

POST /v1/voice-clone/

Upload 6–30 seconds of clean reference audio as multipart/form-data. On success the response includes a voice_id that can be passed straight to TTS. Returns HTTP 201.

Authentication

Pass your API key in the xi-api-key header. See Authentication.

Form fields

Field	Type	Required	Description
`audio`	file	Yes	Reference audio (WAV, MP3, WebM). 6–30 seconds recommended. Maximum 50 MB.
`name`	string	No	Display name for the cloned voice. Defaults to "Cloned Voice". Truncated to 100 characters.

Response

json

{
  "voice_id": "clone-9f3a1c4b",
  "name": "My Brand Voice",
  "status": "ready"
}

Examples

bash

curl -X POST https://sauti.finiflowlabs.com/v1/voice-clone/ \
  -H "xi-api-key: YOUR_KEY" \
  -F "audio=@reference.wav;type=audio/wav" \
  -F "name=My Brand Voice"

python

import requests

with open("reference.wav", "rb") as f:
    response = requests.post(
        "https://sauti.finiflowlabs.com/v1/voice-clone/",
        headers={"xi-api-key": "YOUR_KEY"},
        files={"audio": ("reference.wav", f, "audio/wav")},
        data={"name": "My Brand Voice"},
    )

response.raise_for_status()
print(response.json())

Limits & errors

413 — audio file exceeds 50 MB.
422 — invalid or too-short audio.
429 — rate limit exceeded. See Rate Limits.

Try it interactively in the Voice Clone playground.