Skip to content

About

Turn lecture recordings into transcripts, summaries, and structured study notes — local Whisper + OpenRouter. macOS (Apple Silicon).

Resources

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

27 Commits

Folders and files

Repository files navigation

Note Pilot

Turn lecture recordings into transcripts, summaries, and structured study notes — on your own Mac.

Note Pilot is a desktop app for students. Drop in a recording of a lecture and it:

  1. Transcribes it locally with Whisper (large-v3-turbo, Metal-accelerated) — your audio never leaves your computer.
  2. Generates a concise Summary and detailed Structured Notes via a free OpenRouter model.
  3. Keeps everything in a course-organized library and exports to PDF or Word (.docx).

Built for Spanish + English lectures. Optimized for Apple Silicon Macs.


Requirements

  • Apple Silicon Mac (M1 or newer). Intel Macs are not supported yet.
  • macOS 12 or later.
  • A free OpenRouter API key (for summaries/notes — see below). Transcription works fully offline without it.
  • ~2 GB free disk space (the Whisper model downloads on first use).

Install

Option A — Download the app (recommended)

  1. Go to the Releases page and download the latest Note Pilot.dmg.
  2. Open the DMG and drag Note Pilot to Applications.
  3. Launch it — the app is signed and notarized by Apple, so it opens normally with no security warnings.
  4. On first transcription, Note Pilot downloads the Whisper model (~1.5 GB) and the speech-detection components. This is one-time.

Updates are automatic. When a new version is released, the app downloads it in the background and shows a "Restart to update" banner — no manual re-download.

Option B — Run from source

git clone https://github.com/jarcos/note-pilot.git
cd note-pilot
npm install
npm run rebuild      # compiles the native SQLite module for Electron
npm start

Getting your free OpenRouter API key

Transcription is free and offline. The Summary and Notes features use OpenRouter, which needs a free key. Here's how to get one and add it:

  1. Go to openrouter.ai and click Sign in (you can use Google/GitHub). Creating an account is free and needs no credit card.
  2. Once signed in, open the Keys page: openrouter.ai/keys (your account menu → Keys).
  3. Click Create Key, give it a name like Note Pilot, and click Create.
  4. Copy the key — it looks like sk-or-v1-…. You won't be able to see it again, so copy it now.
  5. In Note Pilot, click ⚙ Settings (top-left), paste the key, and click OK.

That's it — the Generate summary and Generate notes buttons will now work.

About the free tier

Note Pilot defaults to OpenRouter's openrouter/free auto-router, which picks from whatever free models are available (so it keeps working even when individual models change). Free models are rate-limited (roughly a couple hundred requests/day). Each lecture uses ~12–15 requests, so a few lectures a day is fine. If you hit limits often, adding a small one-time credit to your OpenRouter account raises the ceiling substantially.


How to use

  1. Import — drag a lecture audio file (.m4a, .mp3, .wav, …) onto the drop zone, or click to choose. Pick the language (defaults to Spanish).
  2. Transcribe — watch the progress + timer; the transcript appears with timestamps and the lecture is saved to your library.
  3. Organize — create courses (+ Course) and move lectures into them; rename or delete from the lecture toolbar.
  4. Summarize — open the Summary or Notes tab and click Generate.
  5. Export — use the Export buttons to save the current tab (Transcript / Summary / Notes) as PDF or Word.

Privacy

  • Audio is processed entirely on your device — recordings never leave your Mac.
  • Transcripts are sent to OpenRouter when you generate a Summary or Notes. Free-tier AI providers may use submitted text to improve their models. Don't generate summaries for material you consider confidential. (You can skip generation entirely and just use the local transcript.)
  • Your API key is stored locally in the app's data folder (config.json, owner-only permissions) and is never committed or sent anywhere except OpenRouter.

Limitations

  • Apple Silicon only (for now).
  • Summaries/Notes are AI-generated: spot-check important facts against the transcript.

How it works

Electron app. whisper.cpp (Metal) for transcription, ffmpeg for audio decoding, SQLite for the local library, and a map-reduce pipeline over the transcript for Summary/Notes via OpenRouter. Generated content is stored as Markdown and rendered to PDF (Chromium printToPDF) and DOCX (html-to-docx).

License

MIT © José

About

Turn lecture recordings into transcripts, summaries, and structured study notes — local Whisper + OpenRouter. macOS (Apple Silicon).

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages