Migration Guide

rtAV is a drop-in replacement for OpenAI Realtime API with video output capabilities. Migrate your existing code with minimal changes.

Why Migrate?

Video Output: Get real-time video avatars in addition to audio
Custom Faces & Voices: Upload your own custom assets
Competitive Pricing: Transparent $6/hour pricing
Same API: Compatible with existing OpenAI Realtime SDKs

Quick Migration Steps

Step 1: Update API Endpoint

Change the base URL from OpenAI to rtAV:

// Before (OpenAI)

const client = new RealtimeClient('https://api.openai.com/v1/realtime');

// After (rtAV)

const client = new RealtimeClient('https://api.rtav.io/v1/realtime');

Step 2: Update API Key

Replace your OpenAI API key with your rtAV API key:

// Get your API key from platform.rtav.io

const apiKey = 'rta_...'; // Your rtAV API key

Step 3: Enable Video (Optional)

Add video output to your session configuration:

const session = await client.sessions.create({
  model: 'gpt-4',
  voice: 'alloy',
  video: true, // Enable video output
  faceImage: 'https://example.com/face.jpg' // Optional: custom face
});

Code Comparison

OpenAI Realtime

import { RealtimeClient } from '@openai/realtime';

const client = new RealtimeClient({
  apiKey: 'sk-...',
  baseURL: 'https://api.openai.com/v1'
});

const session = await client.sessions.create({
  model: 'gpt-4',
  voice: 'alloy'
});

session.on('audio', (audio) => {
  // Handle audio only
});

rtAV

import { RealtimeClient } from '@openai/realtime';

const client = new RealtimeClient({
  apiKey: 'rta-...',
  baseURL: 'https://api.rtav.io/v1'
});

const session = await client.sessions.create({
  model: 'gpt-4',
  voice: 'alloy',
  video: true // NEW: Video support
});

session.on('audio', (audio) => {
  // Handle audio
});

session.on('video', (video) => {
  // NEW: Handle video
});

API Compatibility

rtAV maintains full compatibility with OpenAI Realtime API endpoints and events:

All standard events (conversation.item.created, response.audio.delta, etc.)
Same WebSocket protocol
Compatible with existing OpenAI Realtime SDKs
Additional video events and capabilities

New Features

Video Output

Get real-time video avatars synchronized with audio responses. Perfect for creating immersive conversational experiences.

Custom Assets

Upload custom faces and voices to create unique avatars that match your brand or use case.

Need Help?

If you encounter any issues during migration, check out our API Reference or contact support.