Follow the audio.
Both pipelines leave the Mac. Only one of them ends with something stored on an account.
Wispr Flow — cloud
- 01Microphone, on your shortcutMenu-bar app, feels instant to open.
- 02Audio uploaded over the internetRequired on every platform — Mac, Windows, iOS, Android.
- 03Transcribed on Wispr’s serversNo connection, no transcript — the app queues the audio instead.
- 04Text returned to your cursorPrivacy Mode can turn off server-side retention, but not the upload.
Four hops. Two of them off your Mac, every single time.
HearFlow — nothing kept by transcription
- 01Microphone, on ⌥SpaceYou hold the shortcut, nothing records in the background.
- 02Audio transcribed, then discardedSent encrypted to api.hearflow.it and on to Groq. The transcription keeps neither the audio nor the text.
- 03Text at your cursorIn the document you already had open.
Three hops, and the transcription keeps nothing. On Windows the transcription runs on your PC.
Both learn
your words.
The questions worth asking.
Does Wispr Flow need an internet connection? Does HearFlow?
Wispr Flow does: its own help centre documents that transcription needs an active connection, and without one the app can only record and queue your audio for later.
HearFlow on Mac also transcribes over the network by default, for accuracy. If the service cannot be reached and the local Whisper model has been downloaded, the app falls back to your Mac — that model is not part of the download, so treat the fallback as a bonus, not a promise. On Windows, transcription always runs on your PC.
Where does my audio go with Wispr Flow, and where with HearFlow?
With Wispr Flow it is uploaded and transcribed on Wispr’s cloud infrastructure, then the text (and, for Notetaker, the recording) is stored on your account. Wispr added an optional Privacy Mode with zero data retention, but it is off by default.
With HearFlow on Mac the audio goes to api.hearflow.it and from there to Groq (United States) for transcription, then comes back as text. The transcription endpoint does not store the audio or the transcript — only usage counters; dictations are kept only if you turn on Audios sync (90 days). On Windows nothing leaves the PC.
Does Wispr Flow have a personal dictionary like HearFlow’s?
Yes, and it is a real feature, not a gap: Wispr Flow learns names and jargon automatically or lets you add them by hand.
The difference is where that dictionary lives. Wispr’s runs as part of their cloud pipeline; HearFlow’s is a file in your user folder, applied on your computer to the text that comes back.
How much does Wispr Flow cost compared with HearFlow?
At the time of checking, Wispr Flow Pro is $12 per user per month billed annually (about $144 a year) or $15 per month billed monthly; Growth/Enterprise starts at $18 per user per month annually or $23 monthly. There is a free plan capped at 2,000 words a week on desktop and 1,000 on iPhone.
HearFlow is €15 a month, €12 a month billed yearly (€144 a year), or €249 once for a lifetime licence. Over three years that one-off licence is a single €249 payment against roughly $432 on Wispr Pro annual billing, more on monthly billing or on the Growth tier.
Is HearFlow a meeting notetaker like Wispr Flow?
No. Wispr Notetaker records meetings, separates speakers and keeps a searchable archive tied to your account — a real product, built for a different job.
HearFlow is live dictation into whatever app has focus, nothing more. If you need meeting notes, Wispr does that job and this page should have saved you a purchase.
What about a comparison with superwhisper instead?
Different competitor, same page structure: read HearFlow vs superwhisper.
Keep the habit.
Lose the bill.
Download, sign in, press ⌥Space and say a name it has never heard. One payment, no seats, no monthly bill to remember to cancel.
Download for MacmacOS 13+ and Windows x64 · 14 days of Pro to try · €249 one-time