HyperVoice for Mac Is Here — Signed, Notarized, and on Apple Silicon
HyperVoice runs on the Mac now. There’s an Apple Silicon .dmg, it’s signed with an Apple Developer ID, it’s notarized by Apple, and it opens the way any normal Mac app does — double-click, click Open once, and you’re dictating.
If you’ve been waiting for this, that’s the whole headline. If you want the details — what’s the same as the Windows and Linux builds, what’s honestly different, and how to get it — read on.
”Signed and notarized” — why it matters
Plenty of indie Mac apps ship unsigned and ask you to run a Terminal command to get past Gatekeeper, or to right-click and choose Open to dodge the “unidentified developer” wall. We didn’t want HyperVoice to be one of those, because asking someone to disable a security check on the very first launch is a bad way to earn trust.
So the Mac build goes through Apple’s full pipeline: it’s code-signed with a Developer ID certificate and submitted to Apple for notarization, which is the automated malware scan that lets macOS vouch for an app it didn’t get from the App Store. The result is the boring, correct experience — the first time you open HyperVoice, macOS shows the standard “this app was downloaded from the Internet, are you sure?” confirmation, you click Open, and every launch after that is a plain double-click. No Terminal, no scary red warning, nothing to disable.
What’s the same as Windows and Linux
The Mac build is the same HyperVoice, not a stripped-down port:
- Local Whisper, on your machine. Speech-to-text runs entirely on-device. Once you’ve downloaded a model, raw dictation works fully offline — no internet required and no audio leaving your Mac. There are 11 Whisper model sizes, from Tiny (~75 MB) up to Large-v3 (~3.1 GB), so you can trade speed for accuracy.
- Metal GPU acceleration. Where the Windows build uses Vulkan, the Mac build uses Apple’s Metal API to run transcription on the GPU, so it stays fast on Apple Silicon even at larger model sizes.
- A global hotkey into any app. Press your hotkey anywhere, speak, and the text is pasted at your cursor — your editor, browser, Slack, Notes, wherever you’re typing.
- 99 languages, your choice of flow. Whisper handles 99 languages, with toggle or hold-to-talk, a custom dictionary for names and jargon, and local history and stats kept on your machine.
The honest caveats
We’d rather you hear these from us than discover them:
- Apple Silicon only. This first build is
arm64— M1, M2, M3, M4. Intel Macs aren’t supported in this release. - Whisper engine only. The optional NVIDIA Parakeet engine is Windows-only, because it runs on DirectML, a Windows API. Whisper covers the same languages and runs GPU-accelerated on Apple Silicon, so you’re not missing dictation features — just that one alternate engine.
- It’s beta. The Mac build hasn’t logged the real-world hours the Windows build has. It’s good, and it’s signed and notarized, but if you hit a rough edge, tell us — most issues get patched within a day or two.
A note on privacy
The privacy line we hold to is specific, so it’s worth being precise here too: the speech engine is local. Your audio is captured, transcribed, and discarded on your own Mac, and that part works with no network at all. The optional AI cleanup modes — tidy up grammar, rewrite as an email, summarize — are a separate, opt-in step that sends the transcribed text (never the audio) to a cloud provider, whether that’s HyperVoice Cloud or your own OpenAI / Anthropic key. Turn cleanup off and the whole thing stays on your machine. We won’t tell you the entire app is offline, because that wouldn’t be true — but the part that hears you is.
How to get it
- Grab the
.dmgfrom your dashboard (or directly from hypervoice.app/api/download/mac). - Open it and drag HyperVoice into Applications.
- Launch it and click Open at the standard macOS confirmation.
- Grant Microphone and Accessibility when prompted — the first lets HyperVoice hear you, the second lets it paste at your cursor and listen for the global hotkey.
The macOS install guide has the full walkthrough with the permission screens if you want it.
Pricing is the same everywhere: a free tier of 500 words a day with no card and no time limit, Lifetime at a one-time $49.99, or Pro at $7.99/month with a 7-day trial.
What’s next
Intel Mac support and an iOS app are the two questions we get most. Intel is a maybe — it depends on demand. iOS is genuinely on the roadmap but isn’t here yet, and we’ll only say it’s ready when it actually is.
For now, if you’re on an Apple Silicon Mac and you’ve wanted local, hotkey-driven dictation that opens without a fight, it’s ready. Try HyperVoice free — 500 words a day, no credit card — and let us know how the Mac build feels.
Frequently asked questions
Does HyperVoice run on macOS now?
Yes. HyperVoice ships a macOS build for Apple Silicon (M1 and later) running macOS 11 Big Sur or newer. It is signed with an Apple Developer ID and notarized by Apple, so it opens with a normal one-click confirmation — no Terminal commands and no workarounds. It is a beta build today, newer than the Windows version.
Is the Mac version signed and notarized?
Yes. The .dmg is code-signed with an Apple Developer ID and notarized by Apple. The first time you open it, macOS shows the standard 'downloaded from the Internet' confirmation — click Open and you're in. There is no 'damaged' or 'unidentified developer' warning to get past.
Does it work on Intel Macs?
Not yet. This first build is Apple Silicon (arm64) only — M1, M2, M3, and M4. Intel Mac support is not in this release.
Is my dictation private on Mac?
Speech-to-text runs entirely on your Mac — your audio never leaves the device and raw dictation works fully offline once a model is downloaded. The optional AI cleanup modes are different: they send the transcribed text, never the audio, to a cloud provider and are opt-in. If you leave cleanup off, nothing leaves your machine.
How do I install it?
Download the .dmg from your dashboard, drag HyperVoice into Applications, open it and click Open at the macOS confirmation, then grant Microphone and Accessibility when prompted. The full walkthrough is in the macOS install guide.
Related posts
Best Voice-to-Text Apps for Windows in 2026
A detailed comparison of the top Windows dictation tools in 2026 — HyperVoice, Wispr Flow, Voicy, Dragon NaturallySpeaking, WhisperTyping, and Windows built-in dictation.
Faster Slack and Teams Messages with HyperVoice
Type less, communicate more. HyperVoice's Chat Message mode turns your spoken words into casual, professional messages for Slack, Teams, and other workplace chat apps.
Filing Bug Reports and Tickets with Your Voice
Stop context-switching to write up tickets. HyperVoice's Ticket / Issue mode turns your spoken description into a structured bug report or task — ready to paste into Jira, Linear, or GitHub.