An AI meeting assistant that works during the call
Most meeting tools send a bot in, then email you a summary. Whisperer does the opposite: it runs on your machine, helps while the conversation is happening, and only afterwards turns it into notes, decisions and tasks.
60 free minutes a month · macOS and Windows · no bot in the call
The answers below are an example. Your own base connects in the app.
…
…
Let's start with Q3 — how did conversion move?
Conversion is up 14% — let me pull up the funnel.
And the cost per lead?
Cost per lead is €12.40 — 9% under plan. Most of the gain came from organic.
Not a video — this is Whisperer itself, and the panel really answers. The microphone starts only when you press it.
During
Three things happen while the meeting is still running
Everything else a meeting tool does can wait until tomorrow. These cannot, and this is the whole reason Whisperer is a desktop application rather than a bot.
- Two streams It hears both sides Your microphone and the system audio of the call, recognised separately and labelled by speaker as the transcript runs.
- Your material It answers from what you know An answer is built from the running transcript, your knowledge base and your profile — not from one isolated question typed into a chat.
- Invisible Only you see it The window stays out of screen sharing, recordings and screenshots. Nothing appears in the participant list.
The map
What the assistant is made of
Each part is a page of its own, because each one answers a different question. Start wherever your problem is.
Mechanism
Why nothing joins your call
This is the structural difference, and almost everything else follows from it. A bot is a participant: it needs an invite, it shows up in the list, and in many companies it needs an administrator to approve the app first.
- Appears in the participant list, and everyone can see it
- Needs the invite, and often an admin to allow the app
- Reads the meeting through the conferencing service
- Works after the fact: you get the summary later
- Nothing joins; the participant list does not change
- No integration with the service, so any of them works
- Reads the meeting through your own system audio
- Works during the call, and keeps the record afterwards
The trade is honest and worth knowing: because Whisperer hears the call the way you hear it, your computer has to be in the call. It cannot attend a meeting you are not on.
Limits
Two things to know before you sign up
Neither is a setting you can flip. If either one rules out your case, better to find that out now than after paying.
A finished recording can be uploaded: the Meetings section takes one file up to 25 MB — mp3, m4a, wav, ogg, flac, aac, plus mp4 and webm, from which the audio track is used. What is missing is bulk upload of a whole archive; and it spends the same minutes a live call does.
The other side is heard through system audio: on macOS that means the Screen Recording permission, on Windows nothing extra beyond the microphone. In a physical room with no online call, the transcript will contain you alone.
Questions
What people ask first
What does an AI meeting assistant actually do?
Two jobs that are easy to confuse. During the call it listens and helps: a running transcript, an answer built from your own material, a translation line if the other side speaks another language. After the call it documents: a map of topics, decisions and action items, and a searchable record. Most tools on the market only do the second job.
Does a bot join my meeting?
No. Whisperer is a desktop application that runs above your screen and hears the call through system audio. The participant list does not change, no one is asked to admit a guest, and there is no app for your administrator to approve.
Which services does it work with?
Any of them: Zoom, Google Meet, Microsoft Teams, Telegram, Discord and anything else that plays the other person through your speakers. There is no per-service integration by design, which is exactly why the list has no end.
Can I upload a finished recording?
Yes. In the Meetings section you pick the file with a button or drag it in — one at a time, up to 25 MB: mp3, m4a, wav, ogg, flac, aac, plus mp4 and webm, from which the audio is taken. From there it goes exactly like a live call: transcript, meeting map, decisions and tasks. A whole archive cannot be uploaded in one go, and in no-logs mode upload is unavailable — nothing is stored there by design.
Will the other side see the assistant?
The window is excluded from screen capture by the operating system, so it does not appear in a shared screen, a recording or a screenshot. That protects against software capture only — a camera pointed at your monitor, or a person next to you, is outside what any software can cover.
What does it cost to try?
The Free plan gives 60 minutes of recognition a month. Start, Pro and Max remove the minute limit; the knowledge base and the advanced prompt modes open on the paid plans.
Try it on one call
Not an important one. Take the next call you have, run Whisperer alongside it, and check two things: whether both sides appear in the transcript, and whether what you get afterwards matches what you remember.