Search docs

Find any page, endpoint, or guide.

Sign inSign up

Quickstart

Get started with VoxRouter.

Alpha

VoxRouter is one API for text-to-speech across every major provider. It is OpenAI-compatible, so any OpenAI client works against it with a base URL change. For VoxRouter-aware code, use the first-party SDK — the import line is where provider-specific features land (catalog typing, attribution headers, future routing hints).

Base URL
api.voxrouter.ai/v1
Auth
Bearer pk_…
Status
Alpha

Generate speech (curl)

POST /v1/audio/speech with an input string and a voice. The response body is the raw audio — mp3 by default.

bash
curl https://api.voxrouter.ai/v1/audio/speech \
  -H "Authorization: Bearer $VOXROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "elevenlabs/eleven_turbo_v2_5",
    "voice": "EXAVITQu4vr4xnSDxMaL",
    "input": "Hello from VoxRouter.",
    "response_format": "mp3"
  }' \
  --output hello.mp3

Generate speech (fetch)

Same call from the browser or Node. The response body is a Blob you can play, save, or stream.

typescript
const resp = await fetch("https://api.voxrouter.ai/v1/audio/speech", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${apiKey}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "elevenlabs/eleven_turbo_v2_5",
    voice: "EXAVITQu4vr4xnSDxMaL",
    input: "Hello from VoxRouter.",
    response_format: "mp3",
  }),
});

const audio = new Audio(URL.createObjectURL(await resp.blob()));
audio.play();

List voices

GET /v1/voices returns the catalog for a configured provider. Filter by provider, language, or gender.

bash
curl "https://api.voxrouter.ai/v1/voices?provider=elevenlabs" \
  -H "Authorization: Bearer $VOXROUTER_API_KEY"

OpenAI-compatible fallback

The OpenAI SDK works unchanged against VoxRouter — point its base URL at https://api.voxrouter.ai/v1 and pass your VoxRouter key. The first-party SDK is still the primary path for VoxRouter-aware code.

typescript
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.voxrouter.ai/v1",
  apiKey: process.env.VOXROUTER_API_KEY!,
});

const mp3 = await client.audio.speech.create({
  model: "elevenlabs/eleven_turbo_v2_5",
  voice: "EXAVITQu4vr4xnSDxMaL",
  input: "Hello from VoxRouter.",
});

Current scope

  • TTS only. Full voice-agents routing returns later.
  • Alpha — endpoints and response shapes may still change.
  • Keep API keys out of source control; inject at runtime.