Realtime API

const API_KEY = "sk_YOUR_KEY";
const WS_URL = `wss://relay.addisassistant.com/ws?apiKey=${encodeURIComponent(API_KEY)}`;
const INPUT_RATE = 16000;
const OUTPUT_RATE = 24000;

let socket;
let inputContext;
let outputContext;
let stream;
let source;
let processor;
let canStreamAudio = false;
let nextStartTime = 0;
const playing = new Set();

async function startRealtime() {
  // Output context for AI playback
  outputContext = new (window.AudioContext || window.webkitAudioContext)({
    sampleRate: OUTPUT_RATE,
  });
  await outputContext.resume();
  nextStartTime = outputContext.currentTime;

  // Input context for mic capture
  inputContext = new (window.AudioContext || window.webkitAudioContext)({
    sampleRate: INPUT_RATE,
  });
  await inputContext.resume();

  stream = await navigator.mediaDevices.getUserMedia({ audio: true, video: false });
  source = inputContext.createMediaStreamSource(stream);
  processor = inputContext.createScriptProcessor(2048, 1, 1);

  // Connect processor through a muted gain so callback keeps firing without mic echo
  const mute = inputContext.createGain();
  mute.gain.value = 0;
  source.connect(processor);
  processor.connect(mute);
  mute.connect(inputContext.destination);

  processor.onaudioprocess = (event) => {
    if (!canStreamAudio || socket?.readyState !== WebSocket.OPEN) return;

    // Browser gives Float32 (-1..1), convert to PCM16 and send as base64 JSON envelope
    const float32 = event.inputBuffer.getChannelData(0);
    const int16 = floatTo16BitPCM(float32);

    socket.send(
      JSON.stringify({
        data: arrayBufferToBase64(int16.buffer),
        mimeType: "audio/pcm;rate=16000",
      }),
    );
  };

  socket = new WebSocket(WS_URL);

  socket.onopen = () => {
    console.log("Connected. Waiting for the ready event...");
  };

  socket.onmessage = async (event) => {
    const message = JSON.parse(event.data);

    switch (message.type) {
      case "ready":
        canStreamAudio = true;
        break;
      case "audio":
        await playPcm16Base64(message.data, message.sample_rate || OUTPUT_RATE);
        break;
      case "interrupted":
        stopPlayback(); // the user spoke over the assistant: drop queued audio
        break;
      case "transcript":
        if (message.finished) console.log(`${message.role}: ${message.text}`);
        break;
      case "turn_complete":
        if (message.usage) console.log("Usage:", message.usage);
        break;
      case "warning":
      case "session_ending":
        console.warn(message);
        break;
      case "error":
        console.error("Realtime error:", message.error);
        break;
    }
  };

  socket.onerror = (event) => {
    console.error("WebSocket error:", event);
  };

  socket.onclose = (event) => {
    console.log(`WebSocket closed code=${event.code} reason=${event.reason || "n/a"}`);
  };
}

function stopRealtime() {
  canStreamAudio = false;

  if (socket && (socket.readyState === WebSocket.OPEN || socket.readyState === WebSocket.CONNECTING)) {
    socket.close(1000, "client-stop");
  }

  if (processor) processor.disconnect();
  if (source) source.disconnect();
  if (stream) stream.getTracks().forEach((t) => t.stop());

  if (inputContext) inputContext.close();
  if (outputContext) outputContext.close();
}

function floatTo16BitPCM(float32Array) {
  const int16Array = new Int16Array(float32Array.length);
  for (let i = 0; i < float32Array.length; i++) {
    const s = Math.max(-1, Math.min(1, float32Array[i]));
    int16Array[i] = s < 0 ? s * 0x8000 : s * 0x7fff;
  }
  return int16Array;
}

function arrayBufferToBase64(arrayBuffer) {
  const bytes = new Uint8Array(arrayBuffer);
  let binary = "";
  for (let i = 0; i < bytes.length; i++) {
    binary += String.fromCharCode(bytes[i]);
  }
  return btoa(binary);
}

function base64ToUint8Array(base64) {
  const binary = atob(base64);
  const bytes = new Uint8Array(binary.length);
  for (let i = 0; i < binary.length; i++) {
    bytes[i] = binary.charCodeAt(i);
  }
  return bytes;
}

async function playPcm16Base64(base64Audio, sampleRate = 24000) {
  const bytes = base64ToUint8Array(base64Audio);
  const pcm16 = new Int16Array(bytes.buffer);
  const float32 = new Float32Array(pcm16.length);

  for (let i = 0; i < pcm16.length; i++) {
    float32[i] = pcm16[i] / 32768;
  }

  const buffer = outputContext.createBuffer(1, float32.length, sampleRate);
  buffer.copyToChannel(float32, 0);

  const src = outputContext.createBufferSource();
  src.buffer = buffer;
  src.connect(outputContext.destination);
  playing.add(src);
  src.onended = () => playing.delete(src);

  const startAt = Math.max(outputContext.currentTime, nextStartTime);
  src.start(startAt);
  nextStartTime = startAt + buffer.duration;
}

function stopPlayback() {
  for (const src of playing) src.stop();
  playing.clear();
  nextStartTime = outputContext.currentTime;
}

import asyncio
import base64
import json
import pyaudio
import websockets

# Audio Config
FORMAT = pyaudio.paInt16
CHANNELS = 1
RATE = 16000
CHUNK = 1024  # ~64ms

async def realtime_chat():
    api_key = "sk_YOUR_KEY"
    uri = f"wss://relay.addisassistant.com/ws?apiKey={api_key}"
    ready_event = asyncio.Event()
    
    async with websockets.connect(uri) as websocket:
        # 1. Setup Microphone
        p = pyaudio.PyAudio()
        stream = p.open(
            format=FORMAT,
            channels=CHANNELS,
            rate=RATE,
            input=True,
            frames_per_buffer=CHUNK,
        )

        async def receive_messages():
            async for raw in websocket:
                message = json.loads(raw)
                if message.get("type") != "audio":
                    print("<", message)

                if message.get("type") == "ready":
                    ready_event.set()
                elif message.get("type") == "error":
                    print("Realtime error:", message["error"])
                    break

        print("Listening... (Press Ctrl+C to stop)")

        try:
            receive_task = asyncio.create_task(receive_messages())
            await ready_event.wait()

            while True:
                # Read raw PCM16 bytes from mic and send as base64 JSON message
                data = stream.read(CHUNK)
                await websocket.send(
                    json.dumps(
                        {
                            "data": base64.b64encode(data).decode("utf-8"),
                            "mimeType": "audio/pcm;rate=16000",
                        }
                    )
                )
                
        except KeyboardInterrupt:
            pass
        finally:
            receive_task.cancel()
            stream.stop_stream()
            stream.close()
            p.terminate()

asyncio.run(realtime_chat())