Live meeting transcription
Choose your meeting audio. Get a transcript, summary and action items when you finish.
Checking online transcription availability…
Email my notes (optional)
Add it now or after the meeting. Confirm once for automatic delivery after meetings of 150+ words.
Tell the other participants before starting. · Recording guidance
Transcription settings and audio help
Saved recording transcriptsChrome extension
Already have a transcript? Paste it and get the brief
From a call recorder, your phone’s own call transcript, Zoom or Teams — any text works. Only the text is sent, to write the brief.
Your meetings on this device (0)
Reopen a transcript or brief here. Saved in this browser only; download important meetings.
Looking for recording transcripts or subtitles? Open saved recording results →
About saved results
Clearing site data removes these meetings. Safari may also clear them after about a week without a visit. Downloaded copies and emails you received stay outside this browser archive.
Loading saved meetings…
How it works
- Open Google Meet, Zoom on the web or Teams on the web. Tell the other participants you will be taking notes.
- Press Start transcription, choose the meeting tab in the browser sharing box and leave Also share tab audio on. Press Share, then allow the microphone to include your voice. Keep this page open during the meeting.
- Press Stop and finish. Meetings of 150 or more words get a summary and action items. If capture was interrupted, review the transcript first. Confirm your email before or after the meeting to receive the notes with the full transcript attached.
If the transcript stays empty
If the timer runs, the meeting sound meter sits flat and no text appears, the page says “We can’t hear your meeting” after about 20 seconds. Check what you shared first, then work down this list.
Nothing you say is in the transcript, but everyone else is
This one is not a fault — it is how shared audio works. Sharing a tab or a screen gives you what your computer is playing, which is the other participants. It never contains your own voice, because a meeting does not play your voice back to you. Your side of the conversation reaches the transcript only through your microphone.
So if your own words are missing, the microphone was refused, muted or set to the wrong input. Watch the second meter while you talk: it turns green when your microphone is genuinely being heard. The page asks you to share the meeting first, then requests microphone access if you chose to include your voice.
On laptop speakers the microphone also hears the meeting. With Chrome’s on-device recognition, the page removes that echo from your side of the transcript, but the echo can drown out what you say at the same moment — headphones give the cleanest record of your own words.
You shared a screen or a window instead of a tab
A window or a whole screen shares sound differently from a tab, or not at all, depending on your computer and on the options ticked in the sharing box. When your meeting is in this same browser, sharing that tab is the cleanest choice: it captures the call and nothing else your computer happens to be playing.
Your meeting is in another browser or app and no sound comes through
Choose Another browser or app before you start, if it is offered. Pick the window of the browser or app running your meeting, or your entire screen when requested. Enable its audio switch before you press Share. If the window has no audio, try Use computer audio instead where offered: it asks for the entire screen with sound, which may include other apps. On a Mac, your browser may first need permission: System Settings → Privacy & Security → Screen & System Audio Recording (called Screen Recording on some Macs), then quit and reopen the browser. Whether that sound comes through depends on your computer and browser, so try it once before an important meeting. If there is no audio checkbox or it stays silent, open the meeting in the same Chrome or Edge browser as Meetings Brief and choose Tab in this browser, or transcribe a recording.
The tab audio switch was clicked off
Also share tab audio sits under the list of tabs in the sharing box and is normally on already. If it was clicked off, the tab is shared without its sound. Press Try again and leave it on.
You shared the wrong tab
The pictures in the sharing box are small and one meeting tab looks much like another. Go to your meeting tab and click Share this tab instead, which Chrome shows at the top of it. If you don’t see it, or the meeting meter still stays flat, press Stop and start again, and pick the tab your meeting is in.
Nobody has spoken yet, or the first segment is not done
With Chrome’s on-device recognition, words appear a second or so after someone speaks; the very first time, Chrome downloads the language once before it starts. With Whisper, text arrives in 30-second segments, so the first one takes at least that long — plus a one-off wait while the speech model downloads. Silence produces no words, which is correct behavior rather than a fault.
You are not in Chrome or Edge on a computer
For meeting-tab audio, open Meetings Brief and the meeting in the same desktop Chrome or Edge browser. Choose Tab in this browser and leave tab audio on. A browser without shared-audio support cannot gain it from a setting on this page. Microphone-only mode can hear a conversation in the room or a call playing on another device’s speaker. You can also use the free transcriber on an existing recording. Phones still cannot capture a call running on that same phone.
The meeting is muted on your machine
Tab capture takes the audio the tab is playing. If you muted the tab in Chrome, there is nothing to take. Your own microphone is separate and is on unless you switch it off before starting.
How this differs from Otter, Fireflies and Google’s own transcripts
Two common ways of transcribing a Google Meet call are a bot that joins the meeting, or the transcript built into paid Workspace plans. This takes another route — your own browser listens to a tab you are already watching.
| This page | A meeting bot (such as Otter’s Notetaker, Fireflies or tl;dv) | Google Meet’s own transcript | |
|---|---|---|---|
| Something joins the call | No | Yes — the bot is a participant | No |
| Who turns the audio into text | Your own computer — the audio is not uploaded | The bot’s service | |
| Cost | Free, no paid plan | Free plans with limits, and paid plans | Eligible paid Workspace editions, including Workspace Individual on a personal Google Account |
| Labels who spoke | You and others, in Chrome | Yes | Yes |
| Works in any browser | Tab audio: Chrome or Edge on a computer. Microphone: any | Yes | Yes |
Google’s native transcripts work on computers and Android devices with an eligible plan. A free personal account does not include them, but Workspace Individual adds them to a personal Google Account. See Google’s transcript requirements.
Not joining the call does not always mean the audio stays on your computer. Otter’s desktop app can record a Zoom meeting without its Notetaker joining, but Otter’s own help says the app uploads the audio to make the transcript (checked 7 Oct 2026; sources in Otter vs Zoom). Fireflies’ Chrome extension also captures Google Meet without its bot. The table compares capture modes, not every mode a vendor offers.
The last two rows are the honest cost of the approach and we are not going to bury them. Telling the other participants apart needs a second model that separates voices, which is not running here, so at best you get your own words marked apart from everyone else’s. And tab audio only exists in Chrome and Edge on a computer; elsewhere the page can only listen through the microphone. If either of those matters more to you than keeping the audio on your own machine, a bot-based tool is the better choice — our Otter alternatives and Fireflies alternatives guides set out what each free tier actually caps.
For why Google’s built-in transcript is missing from most accounts, and which Workspace tiers include it, see how to transcribe a Google Meet call free.
What you get when the meeting ends
Press Stop and finish and the page moves to your brief, with the full transcript below it, ready to copy or download as a text file. The transcript is also saved in this browser as it goes and listed under Your meetings on this device, so you can open it again later on the same computer.
A meeting with more than about 150 words of conversation also gets a brief, written as soon as you stop: what the meeting was for and what came of it, the decisions, the action items with whoever was named against them, the questions left open and a follow-up email ready to send. It appears on this page. If capture was interrupted, automatic summaries and email pause so you can review the transcript first.
Confirm your email address once — we send a 6-digit code — and from then on the brief is emailed to you when you press Stop: on a computer after a 10-second countdown you can cancel, and on a phone with no countdown, because a phone is often locked the moment the meeting ends. The full transcript comes attached as a text file, because that is the part you will want to search in three months. A long meeting takes a few minutes to write, and the page tells you when it has actually been sent rather than only that it was queued.
The brief goes to the single address you confirmed — not to everyone who attended. Shared tab audio carries the sound of the meeting and no list of who is in the room, so there is no attendee list to send to. Forwarding it is a deliberate act on your part, which for a recording of other people is the right default.
Accuracy, languages and the size of the download
In Chrome on Windows, Mac or Linux, this uses Chrome’s own on-device speech recognition for English, Hindi, Spanish, French, German, Italian, Portuguese, Russian, Polish, Turkish, Chinese (Simplified), Chinese (Traditional), Japanese, Korean, Indonesian, Vietnamese and Thai, when Chrome offers that language on your computer — the page checks before you start. Chrome downloads each language once. In our English and Hindi tests, words appeared about a second after they were spoken, and on our 45-second, two-voice English test recording it got 2 of 119 words wrong; accuracy varies by language and by how clearly people speak.
For the other languages, in other browsers, or if you choose it under the options, the speech model is Whisper, running in your browser through WebGPU where your machine supports it and WebAssembly where it does not. The first meeting downloads about 73 MB of model and about 23 MB of speech engine. When the browser keeps those downloads, later meetings avoid downloading them again; the engine still needs to start. The speech engine is served from this site, but the model itself downloads from Hugging Face the first time, so on a network that blocks Hugging Face, live notes cannot start until that download is allowed. WebGPU can be faster on supported hardware. If recognition falls behind the call, the page shows roughly how many minutes behind it is; it catches up after you stop.
Expect a good transcript of clear speech and an imperfect one of everything else. Crosstalk, heavy background noise, a bad connection on someone’s end and unusual proper nouns all cost accuracy — names of people and companies are the most common thing it gets wrong. Read the brief before acting on it, particularly the part that says who agreed to what.
It handles 21 languages: English, Hindi, Tamil, Telugu, Marathi, Spanish, French, German, Italian, Portuguese, Russian, Polish, Turkish, Arabic, Chinese (Simplified), Chinese (Traditional), Japanese, Korean, Indonesian, Vietnamese and Thai. Telugu, Marathi and Arabic are marked low accuracy: they always use Whisper, which gets many of their words wrong. Hindi and Tamil get the same note whenever Whisper is what will run. It starts on your browser’s language when it is one of these and does not detect the spoken language for you, so choose the meeting’s language next to the Start button before you start; the page remembers your choice. If the meeting switches between languages, choose the one spoken most.
On telling people: announce it at the start. Recording law varies by where each participant is sitting, and some places require every party to agree rather than just one — is it legal to record meetings goes through how the rules differ.
Questions people ask
Does a bot join my meeting?
No. Nothing joins the call: your own browser listens to the meeting tab you are already in, so there is no bot to admit. No bot also means nobody is told automatically: tell the other participants at the start that you are taking notes — the page reminds you, and can copy a short notice for the meeting chat.
Is the audio uploaded?
For live capture in On this device mode, no. Speech recognition runs on your machine — through Chrome's built-in on-device recognition for the languages it covers, otherwise through a Whisper model downloaded into your browser. A language or model download may be needed first. Online live capture instead sends short audio batches to Groq, named in the processing notice before capture; it needs a connection and shared capacity. Requesting an AI brief separately sends transcript text for processing, including the automatic brief after an eligible completed meeting. Existing recordings processed Online on the recording transcription page send audio to the provider named before upload (Cloudflare Workers AI or Groq); read the processing notice first.
Which browsers does this work in?
In desktop Chrome or Edge, press Start transcription and choose the meeting in the browser sharing box. For a browser meeting, select its tab and leave tab audio on; the tab may be in another window of that same browser. For a meeting in another browser or a desktop app, select its window or your screen with audio enabled, where offered. Window and screen audio depend on your computer and browser. If there is no sound checkbox or no sound comes through, move the meeting into the same Chrome or Edge browser as this page, or transcribe a recording afterwards. Microphone capture is also available for in-person conversations and calls playing on another device's speaker. Phones cannot capture a call taking place on that same phone.
Do I need to tell people I am transcribing?
Almost certainly yes, and in some places legally. Recording law varies: some jurisdictions require only one participant to consent, others require every participant — California, where we are, is an all-party-consent state. Your employer's policy may be stricter than the law. Announce it at the start and do not continue over an objection.
What happens if I close the tab?
Capture stops, because it is your browser doing the work, so while a meeting is running the page asks before you leave it. The transcript is saved in this browser every few seconds, so what had been transcribed is kept. Open this page again within a day, in the same browser, and it offers to write the brief of your most recent meeting, if it ran to more than about 150 words. The meeting also stays under “Your meetings on this device”. Saved meetings stay in that browser only; Safari may clear them if you do not visit for about a week, so keep the emailed copy or use Download all my meetings.
Does it label who said what?
Partly. When Chrome's on-device recognition is running, your microphone and the meeting are recognized separately, so the transcript marks what You said apart from what Others said. It does not tell the other participants apart from each other — that needs a separate voice-separation model, which is not running here — and with Whisper the output is continuous text.
Can it transcribe a WhatsApp call or a phone call?
Not on the phone the call is on: a web page cannot hear a call taking place on that same phone, iPhone or Android. Two routes work today. Put the call on speaker and press “Use the microphone only” on a second device, such as a laptop next to the phone. Or, if your phone records or transcribes calls itself, paste that transcript into this page and you get the same brief.
Does it work for Zoom and Teams too?
For Zoom or Teams on the web, press Start transcription and select the meeting tab in the browser sharing box with tab audio on. It can be in another window of the same desktop Chrome or Edge browser. For a meeting in a different browser or the desktop app, try sharing its window or your screen with audio enabled, where offered. Confirm that the Meeting sound meter moves while another participant speaks. If there is no sound option or no audio arrives, open the meeting in the same browser as Meetings Brief, or transcribe a recording afterwards.