Records both sides of any call on your Mac and writes it down. Nothing uploaded, nobody else in the room.
$99 once, never a subscription. Apple silicon, macOS 13 or later.
Instead, you get monthly fees, a bot in the room, transcript add-ons, storage tiers, and your clients’ voices sitting on someone else’s server.
All just to press record.
Start recording from your menu bar. No setup, no meeting bot, no links to paste.
Capture your screen, both sides of the conversation, and a speaker-labeled transcript all at once.
Everything saves as standard files in a folder you choose on your Mac. Open them anywhere, move them, or delete them anytime.
No integrations.
No supported-apps list to check.
It records whatever your Mac is playing.
Every meeting recorder claims privacy. These four you can check yourself, in about a minute, without taking our word for any of it.
Once the speech model is downloaded, HearDog records and transcribes normally with no internet connection.
No login. No password. No cloud library tied to you, your meetings or your data.
Speech recognition runs on your own machine, not on a server. No network required.
HearDog records what your Mac is already playing. No extra participant, no meeting bot, no third party in the room.
One payment, and then it is yours. No monthly fee, no per-seat pricing, no add-on for the transcript, no tier for the feature that was the point.
| Bot notetakers | Built into the app | ||
|---|---|---|---|
| Joins the meeting as a guest | No | Yes, with a name | No |
| Can be blocked by their IT | Nothing to block | Often | Admin controlled |
| Works on a phone call or a webinar | Yes, any audio | No | No |
| A cap on how much you record | Only your disk | Minutes limits by tier | Yes |
| You get files, not a web app | Yes | Export if you pay | Sometimes |
| Decides what your meeting meant | Never, that is your job | Yes, whether you asked or not | Yes |
| What it costs | $99, one time | $10 to $30 per person, every month |
Folded into a bigger subscription |
Adopt, don’t subscribe. Competitor rates at the time of writing.
No. There is no account and no server. The speech recognition runs on your Mac using a model that is downloaded once and never talks to anyone again. That download is the only network request it ever makes. After it, you can turn your Wi-Fi off and record and transcribe exactly as normal.
It transcribes with Whisper, which is a speech recognition model, and it runs entirely on your machine. What it does not do is send your meeting anywhere, summarize it, or form an opinion about it. You get the recording and the words. What happens next is your business.
No. Nothing joins. The far end is heard through your Mac's own audio, the same way you hear it, so there is nothing in the participant list to explain.
Tell them. It is the polite thing, and in some places it is required. Rules differ by state and by country, so it is worth knowing the ones where you and the other person are.
A folder, named by date. An audio file of the whole conversation, a plain text transcript, and the screen recording if you asked for one. Ordinary files you can open in anything.
Around 92% of words on a realistic hour of conversation, and you can teach it the names it keeps getting wrong. An hour takes about two minutes to transcribe.
A Mac with Apple silicon, running macOS 13 or later. It is a 5 MB app that lives in your menu bar.
$99, one time. No subscription, no per-seat pricing, no add-on for the transcript and no tier for the feature that was the point. It releases in 2027, and the people on the early access list hear first.
HearDog releases in 2027. Join the list and you hear it first, along with everything that happens between now and then.
Your address stays in the yard. We never sell it or share it.