Generate voice preview

POST/voices/preview

Generate a short audio sample of a voice. Returns the audio file itself, not JSON: audio/mpeg, or audio/wav for Gradium voices.

Request

curl -X POST "https://www.getrevox.com/api/voices/preview" \
  -H "Authorization: Bearer $REVOX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "voiceId": "4325f426-c4e0-418e-a0e5-97fcdfcdf8e6",
    "provider": "cartesia",
    "text": "Bonjour, ici Camille du cabinet Acme. Je vous appelle pour confirmer votre rendez-vous de jeudi."
  }' \
  --output response.mp3

import { writeFile } from "node:fs/promises";

const res = await fetch("https://www.getrevox.com/api/voices/preview", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.REVOX_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    voiceId: "4325f426-c4e0-418e-a0e5-97fcdfcdf8e6",
    provider: "cartesia",
    text: "Bonjour, ici Camille du cabinet Acme. Je vous appelle pour confirmer votre rendez-vous de jeudi.",
  }),
});
await writeFile("response.mp3", Buffer.from(await res.arrayBuffer()));

import os

import requests

response = requests.post(
    "https://www.getrevox.com/api/voices/preview",
    headers={"Authorization": f"Bearer {os.environ['REVOX_API_KEY']}"},
    json={
        "voiceId": "4325f426-c4e0-418e-a0e5-97fcdfcdf8e6",
        "provider": "cartesia",
        "text": "Bonjour, ici Camille du cabinet Acme. Je vous appelle pour confirmer votre rendez-vous de jeudi.",
    },
    timeout=30,
)
response.raise_for_status()
with open("response.mp3", "wb") as file:
    file.write(response.content)

Try itSend this request live

Response

Success

// Binary body: audio/mpeg

{
  "error": "Invalid request: a field or parameter failed validation"
}

{ "error": "Server error" }

Request body

application/json

voiceIdstringrequired
The voice ID to generate a preview for.
provider"cartesia" | "elevenlabs" | "gradium"required
The provider to use for the preview.
textstring
Custom text for the preview. If omitted, a short default is used.
at most 2000 characters
assistantPromptstring
Prompt of the assistant the voice is for. When text is omitted, the preview is a few lines that assistant would say on a call.
at most 100000 characters
speednumber
Speed for preview. Cartesia 0.6 to 1.5, ElevenLabs 0.7 to 1.2. Default 1.0.
0.6 to 1.5
volumenumber
Volume for Cartesia preview (0.5 to 2.0). Ignored for other providers.
0.5 to 2
model"sonic-3" | "sonic-3.5" | "sonic-3.6"
Cartesia model for the preview. Defaults to sonic-3.6. Ignored for other providers.

Response

200

The body is the binary file itself (audio/mpeg).

Errors

400Invalid request: a field or parameter failed validation{ "error": string }
500Server error{ "message"?: string, "error"?: string, "bodyErrors"?: {…} }

Error bodies carry an error (or message) string; a 400 adds bodyErrors.issues. See Errors.