OpenAI Realtime API

POST https://88api.ai/v1/realtime/sessions
Content-Type: application/json
Authorization: Bearer $NEW_API_KEY

{
  "model": "gpt-4o-realtime-preview-2024-12-17",
  "voice": "verse"
}

sequenceDiagram
    participant Client
    participant Server
    participant OpenAI

    alt WebRTC Connection
        Client->>Server: Request ephemeral token
        Server->>OpenAI: Create session
        OpenAI-->>Server: Return ephemeral token
        Server-->>Client: Return ephemeral token

        Client->>OpenAI: Create WebRTC offer
        OpenAI-->>Client: Return answer

        Note over Client,OpenAI: Establish WebRTC connection

        Client->>OpenAI: Create data channel
        OpenAI-->>Client: Confirm data channel
    else WebSocket Connection
        Server->>OpenAI: Establish WebSocket connection
        OpenAI-->>Server: Confirm connection

        Note over Server,OpenAI: Begin real-time conversation
    end

async function init() {
  // Get ephemeral key from server - see server code below
  const tokenResponse = await fetch('/session');
  const data = await tokenResponse.json();
  const EPHEMERAL_KEY = data.client_secret.value;

  // Create peer connection
  const pc = new RTCPeerConnection();

  // Set up remote audio from model playback
  const audioEl = document.createElement('audio');
  audioEl.autoplay = true;
  pc.ontrack = (e) => (audioEl.srcObject = e.streams[0]);

  // Add local audio track from browser microphone input
  const ms = await navigator.mediaDevices.getUserMedia({
    audio: true,
  });
  pc.addTrack(ms.getTracks()[0]);

  // Set up data channel for sending and receiving events
  const dc = pc.createDataChannel('oai-events');
  dc.addEventListener('message', (e) => {
    // Receive real-time server events here!
    console.log(e);
  });

  // Start session using Session Description Protocol (SDP)
  const offer = await pc.createOffer();
  await pc.setLocalDescription(offer);

  const baseUrl = 'https://88api.ai/v1/realtime';
  const model = 'gpt-4o-realtime-preview-2024-12-17';
  const sdpResponse = await fetch(`${baseUrl}?model=${model}`, {
    method: 'POST',
    body: offer.sdp,
    headers: {
      Authorization: `Bearer ${EPHEMERAL_KEY}`,
      'Content-Type': 'application/sdp',
    },
  });

  const answer = {
    type: 'answer',
    sdp: await sdpResponse.text(),
  };
  await pc.setRemoteDescription(answer);
}

init();

import express from 'express';

const app = express();

// Create an endpoint for generating ephemeral tokens
// This endpoint works with the client code above
app.get('/session', async (req, res) => {
  const r = await fetch(
    'https://88api.ai/v1/realtime/sessions',
    {
      method: 'POST',
      headers: {
        Authorization: `Bearer ${process.env.NEW_API_KEY}`,
        'Content-Type': 'application/json',
      },
      body: JSON.stringify({
        model: 'gpt-4o-realtime-preview-2024-12-17',
        voice: 'verse',
      }),
    }
  );
  const data = await r.json();

  // Send the JSON received from OpenAI REST API back to client
  res.send(data);
});

app.listen(3000);

// Create data channel from peer connection
const dc = pc.createDataChannel('oai-events');

// Listen for server events on data channel
// Event data needs to be parsed from JSON string
dc.addEventListener('message', (e) => {
  const realtimeEvent = JSON.parse(e.data);
  console.log(realtimeEvent);
});

// Send client event: serialize valid client events to
// JSON and send via data channel
const responseCreate = {
  type: 'response.create',
  response: {
    modalities: ['text'],
    instructions: 'Write a haiku about code',
  },
};
dc.send(JSON.stringify(responseCreate));

import WebSocket from 'ws';

const url =
  'wss://88api.ai/v1/realtime?model=gpt-4o-realtime-preview-2024-12-17';
const ws = new WebSocket(url, {
  headers: {
    Authorization: 'Bearer ' + process.env.NEW_API_KEY,
    'OpenAI-Beta': 'realtime=v1',
  },
});

ws.on('open', function open() {
  console.log('Connected to server.');
});

ws.on('message', function incoming(message) {
  console.log(JSON.parse(message.toString()));
});

# Requires websocket-client library:
# pip install websocket-client

import os
import json
import websocket

NEW_API_KEY = os.environ.get("NEW_API_KEY")

url = "wss://88api.ai/v1/realtime?model=gpt-4o-realtime-preview-2024-12-17"
headers = [
    "Authorization: Bearer " + NEW_API_KEY,
    "OpenAI-Beta: realtime=v1"
]

def on_open(ws):
    print("Connected to server.");

def on_message(ws, message):
    data = json.loads(message)
    print("Received event:", json.dumps(data, indent=2))

ws = websocket.WebSocketApp(
    url,
    header=headers,
    on_open=on_open,
    on_message=on_message,
)

ws.run_forever()

/*
Note: In browser and other client environments, we recommend using WebRTC.
But in Deno and Cloudflare Workers and other browser-like environments,
you can also use the standard WebSocket interface.
*/

const ws = new WebSocket(
  'wss://88api.ai/v1/realtime?model=gpt-4o-realtime-preview-2024-12-17',
  [
    'realtime',
    // Authentication
    'openai-insecure-api-key.' + NEW_API_KEY,
    // Optional
    'openai-organization.' + OPENAI_ORG_ID,
    'openai-project.' + OPENAI_PROJECT_ID,
    // Beta protocol, required
    'openai-beta.realtime-v1',
  ]
);

ws.on('open', function open() {
  console.log('Connected to server.');
});

ws.on('message', function incoming(message) {
  console.log(message.data);
});

// Receive server events
ws.on('message', function incoming(message) {
  // Need to parse message data from JSON
  const serverEvent = JSON.parse(message.data);
  console.log(serverEvent);
});

// Send events, create JSON data structure conforming to client event format
const event = {
  type: 'response.create',
  response: {
    modalities: ['audio', 'text'],
    instructions: 'Give me a haiku about code.',
  },
};
ws.send(JSON.stringify(event));

# Send client events, serialize dictionary to JSON
def on_open(ws):
    print("Connected to server.");

    event = {
        "type": "response.create",
        "response": {
            "modalities": ["text"],
            "instructions": "Please assist the user."
        }
    }
    ws.send(json.dumps(event))

# Receive messages need to parse message payload from JSON
def on_message(ws, message):
    data = json.loads(message)
    print("Received event:", json.dumps(data, indent=2))