Comparison
Pilcrow and VoiceInk
The closest competitor Pilcrow has, and the one with the better answer to the question this site cares most about. Worth reading before you take our word for anything.
Checked 4 September 2026 against tryvoiceink.com.
It is open source. We are not.
Putting this first because it is the most important thing on the page and everything else reads differently once you know it.
Pilcrow’s central claim is that your speech has no way to leave your Mac. You currently have to take that on trust: the app is closed, so the claim rests on our word and on whatever you can observe from outside with Little Snitch or nettop.
VoiceInk publishes its source code. That is not the same kind of claim — it is a stronger one. You can read what it does, and so can anyone else, which means the claim is checkable by people who have no reason to be kind to it. Nothing we write here matches that, and a comparison page that implied otherwise would be exactly the thing this site exists not to be.
If auditability is your deciding factor, stop reading and go and use VoiceInk. That is a real recommendation.
At a glance
| Pilcrow | VoiceInk | |
|---|---|---|
| Source | Closed | Open |
| Price | Free, no licence | $29 once for one Mac, lifetime updates, free trial |
| Minimum macOS | 26 | 14.4 |
| Platforms | macOS | macOS, iOS |
| Speech models | Apple’s, built into macOS | Local models, downloaded and selectable |
| Optional cloud step | None | Cloud Enhancement, off unless enabled — text only, not audio |
| Rewrites your text | No | Yes — modes, and note structuring |
| Track record | Shipped 2026 | 200k+ downloads, 4.9 average rating |
When VoiceInk is the better answer
- You want to verify rather than trust. Covered above, and it is the big one.
- Your Mac is not on macOS 26. VoiceInk needs 14.4, which is years of Macs we cannot serve. If you are on Sonoma or Sequoia this comparison is already over.
- You want it on your iPhone too. VoiceInk is on iOS. Pilcrow is not, and is not planned to be.
- You want the transcript shaped, not just written down. VoiceInk has modes and will structure a rambling note into something with a list and a closing action. Pilcrow has no setting for any of that.
- You want a mature product. 200k+ downloads and a 4.9 average against an app that shipped this year. If something is broken in VoiceInk, someone has probably already hit it.
- You want to choose your model, or trade size against accuracy. Pilcrow gives you no choice at all.
What is actually different
A shorter list than the one above, and it is not about privacy — both transcribe on your Mac, and VoiceInk states plainly that its optional Cloud Enhancement sends transcribed text and never your voice.
- Free, with no licence. VoiceInk has a free trial and then a one-time purchase — currently $29 for a single Mac, with a promotion running when this was checked. That is a fair price for what it does and cheaper than a year of anything subscription-based. But Pilcrow is free with no trial to expire, no key to enter, and no licence step at all, and for some people that is the difference between installing it today and thinking about it.
- Nothing to download after the app. VoiceInk manages local models — you pick one, it lives on your disk. Pilcrow uses the speech models that ship with macOS 26, so the whole app is 3.1 MB and there is no model folder, no size to choose, and nothing to keep current. The cost of that is precisely the version floor above.
- It writes down what you said. No modes, no structuring, no rewriting. Whether that is a feature or a missing one depends entirely on whether you wanted your own phrasing to survive.
What we are not claiming
That Pilcrow is more private than VoiceInk. It is not a claim we can support and we are not making it. Both do their transcription locally, VoiceInk’s optional cloud step is documented and excludes audio, and theirs can be inspected while ours cannot.
The narrow, honest version is this: Pilcrow has no cloud step to enable, optional or otherwise, and no licence to activate. Its one network request asks this site whether a newer version exists, and Settings can switch that off. That is a difference in what exists, not a difference in trustworthiness, and on trustworthiness the open repository wins.
What this comparison does not settle
Accuracy. VoiceInk runs local models on the Neural Engine; Pilcrow runs Apple’s SpeechAnalyzer. We have not benchmarked them, they are different models, and VoiceInk has a free trial while Pilcrow is free outright — testing both on your own voice costs an afternoon and settles it better than either of our websites will.
Whether we should open source Pilcrow. It is a fair question to ask after reading this page, it has not been decided, and pretending otherwise here would be its own kind of dishonesty.
If Pilcrow still sounds like the one
The home page shows how the trigger key works and the privacy policy states the network claim in the form that binds us. The comparison index has the rest, including Superwhisper, which is the other free local option worth your time.
If anything here has gone out of date, or reads as unfair to VoiceInk, we would rather be corrected than leave it standing — support@thealphacircuit.com.