GPT Realtime Whisper

{
  "type": "transcription",
  "audio": {
    "input": {
      "format": {
        "type": "audio/pcm",
        "rate": 24000
      },
      "transcription": {
        "model": "openai/gpt-realtime-whisper"
      },
      "turn_detection": null
    }
  }
}

// Node.js 22+. speech.pcm is 24 kHz, 16-bit, mono PCM; make one with:
// ffmpeg -i speech.mp3 -ar 24000 -ac 1 -f s16le speech.pcm
import { readFile } from "node:fs/promises";

const socket = new WebSocket("wss://api.ohmygpt.com/v1/realtime?intent=transcription", ["realtime", "openai-insecure-api-key." + process.env.OHMYGPT_API_KEY]);
socket.onmessage = (event) => {
  const data = JSON.parse(event.data);
  if (data.type === "conversation.item.input_audio_transcription.delta") process.stdout.write(data.delta);
  if (data.type === "conversation.item.input_audio_transcription.completed" || data.type === "error") {
    console.log("\n" + (data.transcript ?? JSON.stringify(data.error)));
    socket.close();
  }
};
socket.onopen = async () => {
  socket.send(JSON.stringify({ type: "session.update", session: {
    type: "transcription",
    audio: {
      input: {
        format: { type: "audio/pcm", rate: 24000 },
        transcription: { model: "openai/gpt-realtime-whisper" },
        turn_detection: null,
      },
    },
  } }));
  const audio = await readFile("speech.pcm");
  for (let offset = 0; offset < audio.length; offset += 4800)
    socket.send(JSON.stringify({ type: "input_audio_buffer.append", audio: audio.subarray(offset, offset + 4800).toString("base64") }));
  // Streaming transcription models end a turn when you commit it
  socket.send(JSON.stringify({ type: "input_audio_buffer.commit" }));
};