API Reference

OpenAI-compatible text-to-speech at a flat monthly fee. Change one URL, keep your code.

Quickstart

The fastest way to hear something come out of VoicePep is one curl. Grab an API key at voicepep.com/#pricing, then:

curl https://api.voicepep.com/v1/audio/speech \
  -H "Authorization: Bearer vp_live_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "standard",
    "voice": "am_michael",
    "input": "The morning fog lifted off Lake Pepin."
  }' \
  --output hello.mp3

Play hello.mp3. That's the whole product.

Authentication

Every request needs an API key in the Authorization header:

Authorization: Bearer vp_live_your_key

Your key arrives by email after you sign up on a paid tier, or by request at [email protected] for a free tier key. Keys are 32-character hex strings prefixed with vp_live_. Don't commit them to Git.

Swap from OpenAI in one line

If your code already calls OpenAI's /v1/audio/speech endpoint, change one URL and your existing client works:

from openai import OpenAI

client = OpenAI(
    api_key="vp_live_your_key",
    base_url="https://api.voicepep.com/v1",  # <-- one line changed
)

resp = client.audio.speech.create(
    model="standard",           # production renderer today
    voice="am_michael",
    input="Hello from VoicePep.",
)
resp.stream_to_file("out.mp3")

import OpenAI from "openai";
import fs from "fs";

const client = new OpenAI({
  apiKey: "vp_live_your_key",
  baseURL: "https://api.voicepep.com/v1", // <-- one line changed
});

const mp3 = await client.audio.speech.create({
  model: "standard",           // production renderer today
  voice: "am_michael",
  input: "Hello from VoicePep.",
});
const buf = Buffer.from(await mp3.arrayBuffer());
fs.writeFileSync("out.mp3", buf);

POST /v1/audio/speech

Turn text into an audio file.

Request body

Response

Binary audio bytes in the requested format. Content-Type reflects the format (audio/mpeg for mp3, audio/wav for wav, etc.).

Example — full curl

curl -X POST https://api.voicepep.com/v1/audio/speech \
  -H "Authorization: Bearer vp_live_your_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "standard",
    "voice": "bm_george",
    "input": "Standard voice rendering, one line of code.",
    "response_format": "mp3",
    "speed": 1.0
  }' \
  --output voicepep.mp3

GET /v1/quota

Check your usage this month.

curl https://api.voicepep.com/v1/quota \
  -H "Authorization: Bearer vp_live_your_key"

Returns:

{
  "ok": true,
  "valid": true,
  "tier": "walk",
  "monthly_limit": 100000,
  "used_this_month": 12384,
  "remaining_this_month": 87616,
  "resets_at": "2026-08-01T00:00:00Z"
}

POST /v1/checkout

Get a Stripe checkout URL for a paid tier. Used by the pricing page but you can call it too — useful if you want your own signup UI.

curl -X POST https://api.voicepep.com/v1/checkout \
  -H "Content-Type: application/json" \
  -d '{
    "price_id": "price_0TppiGeGMDltLdCvDaBIJ4Ug"
  }'

Returns:

{
  "ok": true,
  "id": "cs_live_...",
  "url": "https://checkout.stripe.com/c/pay/..."
}

GET /health

Simple liveness probe. Returns HTTP 200 with basic service metadata.

curl https://api.voicepep.com/health

Voice list

Twelve voices ship on every paid tier and work with the production standard model today.

Rendering modes

Rate + concurrency limits

Two independent limits per tier: monthly character quota and simultaneous request cap.

Error responses

All error responses are JSON with an error field.

Where to go next

• 108 integrations with copy-paste setup for n8n, iPhone Shortcuts, Home Assistant, and 17 more platforms.
• Listen to all twelve voices before you sign up.
• Pricing table with a Free tier you can start on today.
• Email [email protected] for anything that isn't here.