OpenAI SDK-compatible GPT-Live speech-to-speech

import { readFile, writeFile } from "node:fs/promises";import { setTimeout as sleep } from "node:timers/promises";import OpenAI from "openai";import { LiveWS } from "openai/resources/live/ws";const client = new OpenAI({  apiKey: process.env.ALLMODELS_API_KEY,  baseURL: "https://api.allmodels.io/oai"});const live = new LiveWS(client);const reply = [];let quiet;live.on("session.output_audio.delta", ({ delta }) => reply.push(Buffer.from(delta, "base64")));live.on("session.output_transcript.delta", () => {  // End the session two seconds after the reply's transcript goes quiet.  clearTimeout(quiet);  quiet = setTimeout(() => live.send({ type: "session.close" }), 2000);});live.on("session.closed", async () => {  await writeFile("reply.pcm", Buffer.concat(reply));  live.close();});live.on("error", console.error);live.send({  type: "session.start",  session: {    model: "gpt-live-1", // or "gemini-3.8-live"    audio: { format: { type: "audio/pcm", rate: 24000 }, output: { voice: "marin" } },    delegation: { type: "client" }  }});// Stream 100 ms frames in real time, then two seconds of silence.const audio = Buffer.concat([await readFile("question.pcm"), Buffer.alloc(2 * 24000 * 2)]);for (let offset = 0; offset < audio.length; offset += 4800) {  live.send({ type: "session.input_audio.append", audio: audio.subarray(offset, offset + 4800).toString("base64") });  await sleep(100);}

import asyncioimport base64import osfrom pathlib import Pathfrom openai import AsyncOpenAIasync def stream_audio(connection):    # Stream 100 ms frames in real time, then two seconds of silence.    audio = Path("question.pcm").read_bytes() + bytes(2 * 24000 * 2)    for offset in range(0, len(audio), 4800):        chunk = base64.b64encode(audio[offset:offset + 4800]).decode("ascii")        await connection.session.input_audio.append(audio=chunk)        await asyncio.sleep(0.1)async def main():    client = AsyncOpenAI(        api_key=os.environ["ALLMODELS_API_KEY"],        base_url="https://api.allmodels.io/oai",    )    async with client.live.connect() as connection:        await connection.session.start(session={            "model": "gpt-live-1",  # or "gemini-3.8-live"            "audio": {"format": {"type": "audio/pcm", "rate": 24000}, "output": {"voice": "marin"}},            "delegation": {"type": "client"},        })        sender = asyncio.create_task(stream_audio(connection))        quiet = None        with open("reply.pcm", "wb") as reply:            async for event in connection:                if event.type == "session.output_audio.delta":                    reply.write(base64.b64decode(event.delta))                elif event.type == "session.output_transcript.delta":                    # End the session two seconds after the reply's transcript goes quiet.                    if quiet:                        quiet.cancel()                    quiet = asyncio.get_running_loop().call_later(                        2, lambda: asyncio.ensure_future(connection.session.close())                    )                elif event.type == "session.closed":                    break        sender.cancel()asyncio.run(main())