Skip to content
Textify
Beta macOS · v0.1.1

Voice to text, anywhere on your Mac

Textify Voice is free voice to text software for Mac. Hold the left Option key, say what you mean, let go — and the words appear wherever your cursor already was, in any app. The speech recognition runs on your own machine, so your audio never leaves it, and it keeps working with no internet connection.

Download for Mac (Apple Silicon) 3.1 MB · free · no account

Requires an Apple Silicon Mac (M1 or later), macOS 11 or later. Prefer to convert an existing recording instead? Use the free in-browser audio to text and voice memo to text tools — no download needed.

Read this before you open it: macOS will block the first launch

You will see “Apple could not verify Textify Voice is free of malware that may harm your Mac or compromise your privacy.” The warning is honest — this alpha has no Apple Developer ID and is not notarized, so Gatekeeper has nothing to check it against. Apple Developer enrollment is underway and this goes away once it is signed and notarized.

The dialog gives you only a Done button — there is no “open anyway” in it. Click Done, then go to System Settings → Privacy & Security → scroll to Security → Open Anyway. You do this once.

On macOS 14 Sonoma and earlier you can instead Control-click the app and choose Open. Apple removed that shortcut in macOS 15 Sequoia, so it does nothing on current systems.

Email me when the notarized build ships — no more System Settings dance

Once Textify Voice is signed and notarized, the warning above goes away entirely. Leave your email and we'll send exactly one message when that build is out.

One email when it ships, nothing else.

How to install Textify Voice

  1. 1

    Download and drag it to Applications

    Open the .dmg and drag Textify Voice onto the Applications folder. The disk image window shows these same steps, so you don't have to keep this page open.

  2. 2

    Open it — and expect macOS to refuse

    You'll get “Apple could not verify Textify Voice is free of malware.” That is expected: this alpha isn't notarized yet. The dialog has only a Done button. Click Done, then do step 3.

  3. 3

    System Settings → Privacy & Security → Open Anyway

    Scroll down to the Security section. You'll see “Textify Voice was blocked to protect your Mac” with an Open Anyway button. Click it, authenticate with Touch ID or your password, then click Open Anyway once more. Do this within an hour of the blocked launch — the button disappears after a while, and opening the app again brings it back. You only do this once, ever.

  4. 4

    Allow Microphone and Accessibility

    The app walks you through both. Accessibility is what lets it type into other apps — macOS treats sending keystrokes to another app as privileged, and it's right to.

  5. 5

    Hold the left Option key and talk

    Release when you're done. Your words appear wherever the cursor already was — Slack, Mail, VS Code, a browser text box, anything.

Faster, if you use Terminal. Steps 2 and 3 exist only to clear the “downloaded from the internet” flag. You can do the same thing directly, after dragging the app to Applications: 

xattr -dr com.apple.quarantine "/Applications/Textify Voice.app"

Speech to text that never leaves your Mac

Most dictation software streams your microphone to a server. This does not. A Whisper speech recognition model is downloaded once (148 MB) and everything after that — capturing audio, transcribing it, punctuating it, inserting it — happens on your own hardware, accelerated by your Mac's Neural Engine and GPU. There is no account to create, nothing to upload, and no transcript kept anywhere by us, because we never receive one.

That also means it is genuinely offline voice typing: once the model is on disk, dictation works on a plane, on hotel wifi that hates you, or with the network off entirely.

What you need

macOS
11 Big Sur or later
Chip
Apple Silicon (M1 or later) — Intel Macs are not supported yet
Download
3.1 MB, plus a one-time 148 MB speech model on first launch
Permissions
Microphone, and Accessibility to type into other apps
Languages
English only in this release

Verify what you downloaded

Since you can't lean on Apple's signature here, check the file yourself. Run this against the downloaded .dmg — it should print exactly the string below it. 

shasum -a 256 ~/Downloads/textify-voice-0.1.1.dmg
0a02cd1b72d2740053b7e997f893e0ed8347d3c956ded761a5bfc3e5cd870115

A mismatch means the file was altered or truncated in transit — don't open it. This proves the download matches what we published; it is not a substitute for notarization, which is a statement about the software itself.

Voice to text FAQ

Is this voice to text app free?
Yes — free, with no account, no sign-up, and no subscription. It is an alpha, and we would rather have your bug reports than your money right now.
Does it work offline?
Yes. The speech model downloads once on first launch (about 148 MB) and everything after that runs on your Mac with no internet connection. Dictation works on a plane.
Does my voice get uploaded to a server?
No. Speech recognition runs entirely on your device using a local Whisper model. There is no upload, no account, and no transcript stored on our side. After the first launch the only network request it ever makes is checking whether a newer version exists, and you can turn that off in Settings.
Why does macOS say it cannot check the app for malicious software?
Because it genuinely cannot. This alpha is not yet signed with an Apple Developer ID or notarized, so Gatekeeper has nothing to verify it against. That warning is accurate and you should take it seriously rather than clicking past it out of habit. What we can offer instead is verifiability: the SHA-256 checksum of the exact file we published is on this page, so you can confirm your download matches before opening it. Apple Developer enrollment is in progress; once the app is notarized this step disappears entirely.
How do I open an app macOS has blocked?
On macOS 15 Sequoia and later: open System Settings, go to Privacy & Security, scroll to the Security section, and click Open Anyway next to the blocked app, then confirm. On macOS 14 Sonoma and earlier you can instead Control-click (right-click) the app in Finder and choose Open. Apple removed that second route in Sequoia, which is why the first one is the main path now.
Which Macs does it run on?
Apple Silicon Macs (M1 or later) running macOS 11 Big Sur or later. Intel Macs are not supported yet. Check → About This Mac; if the Chip line mentions Apple, you're set.
How is this different from Apple's built-in dictation?
It runs a Whisper model locally rather than Apple's dictation engine, it works in any app via a single hold-to-talk key, and it applies its own punctuation and formatting pass plus a user dictionary for names and jargon it would otherwise get wrong. It is also honest about being an alpha — Apple's built-in dictation is not going anywhere, and if it already works for you, keep using it.
What are the limitations right now?
English only. Dictation only — voice commands like “delete that” exist in the codebase but are deliberately inert in this build and execute nothing. Apple Silicon only. And accuracy is far better proven on someone dictating a sentence into a text field than on long, messy, multi-speaker recordings.

Deciding between this and something else? See how it compares to Wispr Flow, superwhisper, and VoiceInk — sourced, dated, and including where each one is a better fit than we are.

This is an alpha and it is offered as one. It will have rough edges, and the fastest way to get them fixed is to tell us what happened. For detail, the app can write a diagnostic bundle to a file — run "/Applications/Textify Voice.app/Contents/MacOS/textify-voice" diagnostics in Terminal. It writes locally and sends nothing on its own, so you can read the whole thing before deciding whether to share it.