Skip to main content

L402 API

Pay with Bitcoin, get AI results. Copy-paste the commands below.

Back to all services & pricing

Choose a service:

ImageTextVideoVideo from ImageMusicText to SpeechSpeech to TextVision3D ModelFile ConversionOCRSMSPhone CallDocument AnalysisEmailVoice CloneImage EditPDF MergeHTML to PDFTranslationReceipt OCR

Before You Start

Lightning Wallet

Phoenix, Muun, Wallet of Satoshi, Alby, or any LN wallet

Terminal or Tool

A command line, Postman, or lnget (recommended)

Private & No Signup

No account, no subscription, no personal data. Pay only when you use it

Clone a Voice via L402

Clone a voice from an audio sample. Returns a voice_id that you can use with the Text-to-Speech endpoint to generate speech in that voice. Flat rate: 7,500 sats per clone.

Ideal for creating custom voices for agents, branded TTS, audiobook narration, or any scenario where you need a specific voice without ongoing subscription fees.

API Endpoint: https://sats4ai.com/api/l402/voice-clone

Constraints

  • Audio sample must be a publicly accessible URL or base64 data URI
  • Processing takes 1-5 minutes depending on model and audio length
  • Two quality tiers: speech-02-turbo (faster) and speech-02-hd (higher quality)
  • Accuracy range: 0.0 to 1.0 (default 0.7) — higher values produce closer voice matches but take longer

Step 1: Initial Request (get Invoice)

Send the request with your voice sample URL. You will receive a 402 Payment Required response with a Lightning invoice for 7,500 sats.

Request body:

{
  "voice_file": "https://example.com/my-voice-sample.mp3",
  "model": "speech-02-turbo",
  "accuracy": 0.7
}

Example curl command:

curl -X POST https://sats4ai.com/api/l402/voice-clone \
-H "Content-Type: application/json" \
-d '{"voice_file":"https://example.com/my-voice-sample.mp3","model":"speech-02-turbo","accuracy":0.7}' \
-i

Expected response headers:

www-authenticate: L402 macaroon="<YOUR_MACAROON>", invoice="<LIGHTNING_INVOICE>"

Step 2: Pay the Lightning Invoice

Pay the 7,500-sat invoice with any Lightning wallet (Phoenix, Muun, Alby, Wallet of Satoshi, etc.). Save the preimage your wallet provides after payment.

Step 3: Re-send with Payment Proof

Send the exact same request again with the Authorization header.

The JSON payload must be identical to Step 1.

curl -X POST https://sats4ai.com/api/l402/voice-clone \
-H "Content-Type: application/json" \
-H "Authorization: L402 <YOUR_MACAROON>:<YOUR_PREIMAGE>" \
-d '{"voice_file":"https://example.com/my-voice-sample.mp3","model":"speech-02-turbo","accuracy":0.7}'

Successful response: {"success": true, "voice_id": "abc123", "model": "speech-02-turbo", "message": "Voice cloned successfully! Use voice ID \"abc123\" with the TTS endpoint."}

Easier: Use lnget

lnget POST https://sats4ai.com/api/l402/voice-clone \
  --json '{"voice_file":"https://example.com/sample.mp3","model":"speech-02-turbo"}'

Python Example

import requests

# Step 1 — get invoice
r = requests.post("https://sats4ai.com/api/l402/voice-clone",
    json={"voice_file": "https://example.com/sample.mp3", "model": "speech-02-turbo", "accuracy": 0.7},
    headers={"Content-Type": "application/json"})
# r.status_code == 402
# Parse macaroon and invoice from r.headers["www-authenticate"]

# Step 2 — pay invoice with your Lightning wallet, get preimage

# Step 3 — re-send with auth
r2 = requests.post("https://sats4ai.com/api/l402/voice-clone",
    json={"voice_file": "https://example.com/sample.mp3", "model": "speech-02-turbo", "accuracy": 0.7},
    headers={
        "Content-Type": "application/json",
        "Authorization": "L402 <macaroon>:<preimage>"
    })
print(r2.json())
# {"success": true, "voice_id": "...", "message": "Voice cloned successfully!"}

Request Parameters

voice_file: string required

Public URL to an audio file containing the voice to clone, or a base64 data URI. Supported formats: MP3, WAV, M4A, FLAC.

model: string (optional)

Quality tier. "speech-02-turbo" (faster, default) or "speech-02-hd" (higher quality).

accuracy: number (optional)

Voice matching accuracy. Range: 0.0 to 1.0. Default: 0.7. Higher values produce closer matches but may take longer to process.

need_noise_reduction: boolean (optional)

Apply noise reduction to the voice sample before cloning. Default: false.

need_volume_normalization: boolean (optional)

Normalize volume levels in the voice sample. Default: false.

Use Cases

  • Create a custom branded voice for your agent or product
  • Clone your own voice to generate TTS audio that sounds like you
  • Audiobook narration with a consistent custom voice
  • Accessibility tools with a familiar, personalized voice
  • One-time voice cloning without monthly subscription fees