Screen recorder · AI transcription · Summaries

Record.
Transcribe.
Never forget.

Capture any screen or window. Get a full AI transcript and a AI summary the moment you stop recording.

v0.1.1 · Free with your own API key
I Gave Up
Sessions
Design review
Today · 34 min
Sprint planning
Yesterday · 58 min
1:1 with Sarah
Mon · 22 min
Onboarding call
Mon · 41 min
TL;DR
TL;DR
Key points
Action items (3)
Chapters
Transcript
0:00
Alice: Let's start with the dashboard redesign...
0:24
Bob: I think we should simplify the nav...
1:02
Alice: Agreed. The metrics panel needs a refresh too.

How it works

Four steps. Zero effort.

Hit record. Walk away. Come back to a full summary waiting for you.

01
Pick your source
Choose any window, display, or app. Mic + system audio captured together.
02
Record
Click record. Minimize and forget it. No quality loss, no length limits.
03
AI transcribes
Whisper processes audio in parallel chunks. Full transcript with timestamps.
04
AI summarizes
TL;DR, key points, action items, and chapter markers — ready instantly.

Features

Everything you need, nothing you don't.

Built for people who just want the important parts of a conversation, without the manual work.

Record anything
Full display, individual windows, or a specific app. Mic and system audio captured independently for clean mixing.
Click-to-seek transcript
Every line is timestamped. Click any sentence in the transcript to jump directly to that moment in the video.
AI summaries
TL;DR, key points, action items, and chapter markers generated the moment your recording ends.
Session library
Every recording saved locally with its video, transcript, and summary. Browse and replay anytime.
Parallel transcription
Audio split into chunks and transcribed simultaneously. Long recordings don't mean long waits.
Multiple AI providers
OpenAI Whisper by default. Switch to Deepgram in settings for different speed/accuracy tradeoffs.
Configurable

Demo

See it in action.

A 60-second walkthrough: record, stop, and watch the summary appear.

demo video coming soon


Bring Your Own Key.

Your API key, your data, your bill. No shared backend, no accounts, no hidden usage. Processing goes straight from your machine to the provider you choose.

No sign-ups No hidden usage No vendor lock-in Runs locally

I gave up on finding a tool that did this properly. Most options were browser-only, cloud-dependent, or required accounts for something that should just work in the background.

What's coming

Optional cloud mode

For those who want convenience over control. BYOK will always remain the default — cloud is an add-on, not a requirement.

  • Optional login
  • Sync across devices (PC, Mac, etc.)
  • Secure storage for recordings and transcripts

BYOK will always be free. Cloud will be an opt-in add-on for people who want it.


Changelog & roadmap

What's shipped. What's next.

Built in the open. The roadmap is the plan, and the plan changes based on what people actually use.

Shipped
Screen & window capture
Windows Graphics Capture API · multi-monitor · 30/60 FPS
System + mic audio capture
CPAL · WASAPI loopback · real-time level meters · mute controls
Multi-provider transcription
OpenAI Whisper · Deepgram · BYOK
AI summaries
TL;DR · key points · action items · key decisions · chapters
Session library & synced playback
Click-to-seek · auto-scroll transcript · in-transcript search
Onboarding & settings
3-step setup · auto-transcribe/summarize · multiple storage locations
System tray & auto-updates
Recording indicator · minimize to tray · in-app update check
macOS support
ScreenCaptureKit · Core Audio · full feature parity
Coming soon
Cloud sync & user accounts
Cross-device access · freemium · hybrid BYOK + managed
AssemblyAI provider support
Introduce as a third transcription provider
Global keyboard shortcuts
Start/stop recording via global hotkeys
Offline AI transcription
Local models via whisper.cpp · fully offline operation
Data backup & migration
Migrate local session libraries to cloud sync
Semantic search
AI-powered search across transcripts & summaries
Speaker diarization
Identify different speakers in meeting transcripts
Linux support
PipeWire/PulseAudio · AppImage & .deb

Stop taking manual notes.

Bring your own OpenAI key. No account needed.

v0.1.1 · macOS