Getting Started

Overview

The ORIGIN Neural API gives you access to state-of-the-art voice synthesis powered by four engines: Spark (kokoro — Swift Lane), Flux (f5tts — Clarity Lane), Tide (fish_speech — Clarity Lane), and Echo (moshi — Dialogue Lane). Generate natural speech from text, clone voices, breed new voices with genetic algorithms, and verify ownership on-chain.

All endpoints return JSON unless otherwise noted. Audio endpoints return binary WAV data. The base URL for all API requests is:

https://originneural.ai

Prerequisites

Before you begin, you'll need:

Your First API Call

Let's synthesize speech. Send a POST request to the speak endpoint with your text and a voice ID. The API returns WAV audio data.

curl -X POST https://originneural.ai/api/demo/speak \
  -H "Authorization: Bearer origin_sk_your_key" \
  -H "Content-Type: application/json" \
  -d '{"text": "Hello from ORIGIN Neural!", "voice_id": "default"}' \
  --output speech.wav

Response Format

JSON endpoints return a consistent response shape. Successful requests return a 200 status with the response data. Error responses include an error field with a human-readable message and an HTTP status code.

{
  "voices": [...],
  "total": 42
}

Next Steps

Now that you've made your first API call, explore the rest of the documentation: