← Back to Documentation

API Reference

Complete reference for rtAV API endpoints. Compatible with OpenAI Realtime API.

Base URL

https://api.rtav.io/v1

Authentication

All API requests require authentication using an API key. Include your API key in the Authorization header:

Authorization: Bearer rta_your_api_key_here

WebSocket Endpoint

Create Realtime Session

wss://api.rtav.io/v1/realtime?model=gpt-4

Establishes a WebSocket connection for real-time audio and video streaming.

Query Parameters:

  • model - Model to use (e.g., gpt-4, gpt-3.5-turbo)
  • voice - Voice to use (e.g., alloy, echo, fable)
  • video - Enable video output (true/false)
  • face_id - Custom face ID (optional)
  • voice_id - Custom voice ID (optional)

REST Endpoints

POST /sessions

Create a new session

{ "model": "gpt-4", "voice": "alloy", "video": true, "face_id": "face_123" // optional }

GET /sessions/{id}

Get session information

POST /sessions/{id}/end

End a session

POST /faces

Upload a custom face image

POST /voices

Upload a custom voice

Events

rtAV supports all OpenAI Realtime API events, plus additional video events:

  • conversation.item.created - New conversation item
  • response.audio.delta - Audio chunk
  • response.video.delta - Video chunk (rtAV extension)
  • response.done - Response complete
  • error - Error occurred

Full Documentation

For complete API documentation with all endpoints, parameters, and examples, see our Code Examples page.