Turn a meeting transcript into minutes and a list of who's doing what.

Paste a transcript with speaker names to get decisions, action items with owners and due dates, open questions and a follow-up email, plus a breakdown of who spoke and who got interrupted.

Version 2.0.0 · MIT license · github.com/dguywhoknows/meeting-notes-analyzer

The app itself, running inside this page. Anything you do here is saved in your browser.

Ask it to do things

Press Ctrl+K in the app (Cmd+K on a Mac) and describe what you want. The model chooses from the app's own functions, runs them in order and tells you what changed. Some examples:

What did Tom agree to do, and by when?Looks up Tom's action items and answers.
Mark the analytics task as doneUpdates the action tracker.
Draft the follow-up emailWrites the recap from the minutes.

Instructions you save in the same box, such as "keep it short" or "I'm a beginner", are added to every AI request the app makes.

Features

Transcript parser
Plain 'Name: text', [hh:mm:ss] stamps, WebVTT/SRT with <v> voice tags.
Speaker timeline lanes
Talk-share bars, Gini-coefficient balance indicator.
Dynamics table
Turns, avg/longest turn, questions, interruptions given/received, filler words.
AI minutes grounded
With quotes; never invents owners or dates.
Checkable action items; export Markdown minutes
CSV action items.
Streaming follow-up email draft
Rule-based extraction that works without a model
Commitments ("I'll …"), requests addressed to someone ("Tom, can you …"), decisions and unanswered or parked questions, with owners taken from the speaker or addressee.
Due-date resolution
Weekdays, next <weekday>, tomorrow, next week, end of week/month, "the 21st" and month-day phrases become real dates relative to the meeting.
Minutes page
Editable action items (task, owner, due date, done), decisions with supporting quotes, follow-up email and Markdown export.
Actions page
One tracker across every saved meeting with owner and status filters, overdue highlighting and CSV export.

Runs in your browser

Parsing, timing estimation, speaker analytics, timeline rendering, exports. None of this needs a network connection or a model.

Uses a language model for

Structured minutes extraction with supporting quotes. Follow-up email generation.

Notes

Cost
Free. Each account has a daily allowance of AI requests.
Your data
What you make stays in this browser on this device. Your account is only used for signing in and for the built-in AI. You can export a backup or clear everything from Settings.
AI access
Signed-in users get the built-in AI with nothing to set up. You can also use your own Groq or OpenRouter key under Settings.
Source
The code and tests are on GitHub under the MIT license. The tests also run in your browser.