CASE 06
Meeting Intelligence Copilot
A desktop copilot sits on your machine, hears both sides of the call, and hands you the next line to say in seconds.
- Role
- Design, build, automate
- Year
- 2026
- Stack
- Desktop app · On-device AI
- Status
- In daily use
01 The problem
A hard call moves fast. You are listening, thinking, and choosing your words at the same time, with no one feeding you the right line. You used to do all of that on your own. The right number is somewhere in a file you cannot open mid-sentence, and the perfect answer arrives ten minutes after you hang up. Notes written after the fact do not help in the moment. By then the call is over and the chance is gone.
02 The system
A desktop copilot that sits over the call. It hears your microphone and the other person’s audio at once and writes a live transcript as both of you talk. Press a hotkey and it reads the whole conversation, then streams two replies you can say out loud in seconds. Feed it a deck or your notes first and every line pulls from your real numbers, so you answer with the actual figure instead of a guess. When the call ends, one more hotkey writes a clean summary with action items and saves the full transcript to your notes. The work you used to do alone, and the follow-up you used to scribble afterward, both happen for you while the call is still live.
03 How it holds up
Speech runs on the machine itself with a local model, so the conversation never leaves the laptop. That matters on a sensitive call, where you do not want the audio sent anywhere. The suggestion layer throws out clichés and repeats so it sounds like you, not a chatbot. It is built for speed, because a line that arrives after you have already spoken is useless. It carries 117 tests, and I use it in my own calls every week.
04 The result
A live call comes with a second set of ears and a draft of the next line, grounded in the facts that matter and ready in seconds. You walk in calmer, you answer with real numbers, and the summary is written before you close the laptop. All of it runs on your own machine, so the conversation stays yours.
· How it works
- 01
Launch and set up
Add your keys, pick a speech model, and grant microphone and screen permission once.
- 02
Start the meeting
The app listens to your microphone and the other person's audio at the same time and writes a live transcript.
- 03
Ask for a line
Press a hotkey and it reads the conversation, then streams two ready-to-say replies in your voice.
- 04
Stay grounded
Upload a deck or notes first, and every suggestion pulls from your real numbers, not generic filler.
- 05
Save the meeting
One more hotkey writes a clean summary with action items and the full transcript to your notes.
· Results
- Transcription
- On-device
- Audio
- Both sides
- Tests
- 117
1. Speech runs on the machine with a local model, so meeting audio never leaves the laptop.
2. The suggestion layer rejects clichés and repeats, so it reads like a person, not a chatbot.