> For the complete documentation index, see [llms.txt](https://docs.aimlapi.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.aimlapi.com/api-references/realtime-models/openai/gpt-live-1.md). # GPT-Live {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `openai/gpt-live-1` * `gpt-live-1` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} OpenAI's live voice model for continuous spoken conversations. A GPT-Live session takes a continuous audio stream, is configured once in its first event, and is billed by its duration rather than by tokens. ## Setup your API Key If you don’t have an API key for the AI/ML API yet, feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). ## API Schema The model is served over a WebSocket in the OpenAI GPT-Live protocol. The session flow, client events, billing, limits and errors are described in [Realtime Sessions](/capabilities/realtime-sessions.md#gpt-live). | | | | -------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Connection URL | `wss://api.aimlapi.com/v1/live/sessions` (no `?model=`: the model is named in `session.start`) | | Authentication | Server: `Authorization: Bearer `. Browser: subprotocols `realtime` and `openai-insecure-api-key.` ([client secrets](/capabilities/realtime-sessions.md#browsers-client-secrets)) | | SDK base URL | `https://api.aimlapi.com/v1` | Session options (`session.start`, the first event, sent once): | Option | Values | | -------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `model` | `openai/gpt-live-1` or `gpt-live-1` (required) | | `instructions` | Up to 16,384 characters | | `input` | Up to 128 history messages to seed the conversation, text only, one part each (`developer`, `user` with `input_text`; `assistant` with `output_text`) | | `audio.format` | One format for input and output: `{"type": "audio/pcm", "rate": 24000}`, `{"type": "audio/pcm", "rate": 16000}`, `{"type": "audio/pcmu", "rate": 8000}`, `{"type": "audio/pcma", "rate": 8000}` | | `audio.output.voice` | Built-in voices: `alloy`, `ash`, `ballad`, `beacon`, `bossa`, `brise`, `cedar`, `cinder`, `coral`, `delta`, `echo`, `flitz`, `gleam`, `harema`, `juni`, `marin`, `meridian`, `nira`, `noeul`, `nuri`, `quartz`, `ripple`, `sage`, `shida`, `shimmer`, `sillage`, `stone`, `tempo`, `verse`, `vesper`, `willow` | | `delegation` | `{"type": "client"}` only (applied automatically) | | `store` | `false` only (applied automatically) | ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python # pip install websockets import asyncio import base64 import json import websockets # Insert your AI/ML API key instead of : API_KEY = "" URL = "wss://api.aimlapi.com/v1/live/sessions" # 100 ms of 24 kHz mono PCM16 silence; send real microphone audio instead. CHUNK = base64.b64encode(bytes(4800)).decode() async def stream_audio(ws): for _ in range(100): await ws.send(json.dumps({"type": "session.input_audio.append", "audio": CHUNK})) await asyncio.sleep(0.1) await ws.send(json.dumps({"type": "session.close"})) async def main(): async with websockets.connect( URL, additional_headers={"Authorization": f"Bearer {API_KEY}"} ) as ws: await ws.send(json.dumps({ "type": "session.start", "session": { "model": "openai/gpt-live-1", "instructions": "Greet the user in one short sentence.", "audio": { "format": {"type": "audio/pcm", "rate": 24000}, "output": {"voice": "marin"}, }, }, })) async for message in ws: event = json.loads(message) if event["type"] == "session.started": asyncio.create_task(stream_audio(ws)) elif event["type"] == "session.output_audio.delta": pass # the model's speech: play it elif event["type"] == "session.closed": print("Final usage:", event.get("usage")) elif event["type"] == "error": print("Error:", event["error"]) asyncio.run(main()) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript // npm i ws import WebSocket from "ws"; // Insert your AI/ML API key instead of : const apiKey = ""; const ws = new WebSocket("wss://api.aimlapi.com/v1/live/sessions", { headers: { Authorization: `Bearer ${apiKey}` }, }); ws.on("open", () => { ws.send(JSON.stringify({ type: "session.start", session: { model: "openai/gpt-live-1", instructions: "Greet the user in one short sentence.", audio: { format: { type: "audio/pcm", rate: 24000 }, output: { voice: "marin" }, }, }, })); }); const streamAudio = () => { // 100 ms of 24 kHz mono PCM16 silence; send real microphone audio instead. const chunk = Buffer.alloc(4800).toString("base64"); const timer = setInterval( () => ws.send(JSON.stringify({ type: "session.input_audio.append", audio: chunk })), 100, ); setTimeout(() => { clearInterval(timer); ws.send(JSON.stringify({ type: "session.close" })); }, 10_000); }; ws.on("message", (data) => { const event = JSON.parse(data.toString()); if (event.type === "session.started") { streamAudio(); } else if (event.type === "session.closed") { console.log("Final usage:", event.usage); } else if (event.type === "error") { console.error("Error:", event.error); } }); ws.on("close", (code, reason) => console.log("Closed", code, reason.toString())); ``` {% endcode %} {% endtab %} {% endtabs %}
Response The example streams 10 seconds of audio (silence, in 100 ms chunks), closes the session with `session.close`, and prints the final usage that `session.closed` carries. Values in angle brackets are placeholders. {% code overflow="wrap" %} ``` Final usage: ``` {% endcode %}
With the official OpenAI Node.js SDK, create the client with `baseURL: "https://api.aimlapi.com/v1"` and open the session with its `LiveWS` class; the events are the same. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.aimlapi.com/api-references/realtime-models/openai/gpt-live-1.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.