Live meeting transcript in Python, with who said each line

What it does

This script prints each line of a meeting as it is said, under the name of the person who said it, for as long as LISTEN_FOR allows — then asyncio.wait_for ends it.

It needs a key and a meeting code in MEETFUL_KEY and MEETING. Two curl requests make both. MEETFUL_URL is optional and defaults to https://meetful.com.

Python 3.9 or later. The socket scripts use the websockets package (pip install websockets; this ran with 13.1); the transcript script uses only the standard library.

The code

# Print every line said in a Meetful meeting, with its speaker, as it is said. pip install websockets
# MEETFUL_KEY=tmk_… MEETING=abc-defg-hij LISTEN_FOR=60 python listen.py
import asyncio, json, os
from datetime import datetime, timezone
import websockets

BASE = os.environ.get("MEETFUL_URL", "https://meetful.com")
LISTEN_FOR = float(os.environ.get("LISTEN_FOR", "60"))

async def listen():
    async with websockets.connect(BASE.replace("http", "ws", 1)) as ws:
        await ws.send(json.dumps({"type": "agent-hello", "key": os.environ["MEETFUL_KEY"], "meeting": os.environ["MEETING"]}))
        async for raw in ws:
            m = json.loads(raw)
            if m["type"] == "agent-ready":
                print(f"listening in {m['meeting']}…", flush=True)
            elif m["type"] == "transcript":
                line = m["line"]
                at = datetime.fromtimestamp(line["at"] / 1000, timezone.utc).strftime("%H:%M:%S")
                # The speaker is whose microphone the words came from, not a guess.
                print(f"{at}  {line['speaker']}: {line['text']}", flush=True)
            elif m["type"] == "agent-refused":
                raise SystemExit("Refused: " + m["why"])

async def main():
    try:
        await asyncio.wait_for(listen(), LISTEN_FOR)
    except asyncio.TimeoutError:
        pass

asyncio.run(main())

Run it

MEETFUL_KEY=tmk_… MEETING=abc-defg-hij python listen.py

What it printed

listening in jho-bwva-mia…
14:17:03  Dana: Can we move the launch to the third of March?
14:17:04  Sam: Yes, if the docs are ready by the twentieth.
14:17:05  Dana: Then Sam owns the docs.

Run on 10 November 2026 against a Meetful server built from the same code as meetful.com, with two people — Dana and Sam — in the meeting saying three lines. Nothing above was edited.

How it works

print(..., flush=True) matters when the output goes to a file or a pipe: without it Python holds the lines in a buffer and you see nothing until the script ends.

Each speaker name comes from the microphone the line was spoken into, so a line is never filed under the wrong person. That is the reason to use a meeting's own transcript rather than a recording run through a speaker-guessing model.

What can go wrong

Next

In Python: join a meeting · speak · get the transcript

The same in Node.js · Go

The whole API · Every socket message and webhook · All guides