What is SpeakType?

SpeakType lets you type with your voice. Hold a key, say what you want to write, and let go. Your words appear wherever your cursor is: an email, a chat, a document, a code editor, anywhere.

Most voice typing apps send your recordings to a company's servers and charge a monthly fee. SpeakType does all the work on your own computer, so it's free, it works without internet, and nothing you say ever leaves your computer.

How it works

1. Install SpeakType and open it.
2. Download a speech model when SpeakType asks. This is the part that turns your voice into text. It's a one-time download of 75 MB to 1.6 GB, and SpeakType suggests one that suits your computer.
3. Hold your shortcut and talk. Click into any text box, hold fn on a Mac or Ctrl + Shift + Space on Windows and Linux, speak, and let go. You can change the shortcut in Settings.

Why people use it

• ⚡ Fast. A 35-second ramble becomes neat, punctuated text in under two seconds.
• 🪶 Light. A 15 MB install that stays out of the way while you're not talking.
• ⌨️ Works in every app. It types into whatever you're using, and puts back anything you had copied.
• 🌍 Many languages. Up to 99 languages, depending on the model you pick.
• 📖 Gets your words right. Teach it names and terms it mishears, and filler words like "um" are removed.
• 🕘 Keeps a history. Copy and replay past dictations, and see how much typing you've saved.

Download

Go to the downloads page and, under the newest version, click the file for your computer:

Installing

Mac

1. Open the .dmg file and drag SpeakType into your Applications folder.
2. Open SpeakType. If macOS says it can't check the app, go to System Settings → Privacy & Security, scroll down, and click Open Anyway. You only need to do this once.
3. When asked, allow microphone access (so it can hear you) and accessibility access (so it can type for you).

Windows

1. Run the -setup.exe file.
2. If a blue "Windows protected your PC" box appears, click More info, then Run anyway. You only need to do this once.

Linux

• .deb: double-click it to install with your software center, or run sudo apt install ./SpeakType_*.deb.
• .AppImage: make it runnable with chmod +x SpeakType_*.AppImage, then double-click it.

The extra steps on Mac and Windows are there because SpeakType isn't signed by Apple or Microsoft yet. That's coming soon.

Questions

Is it really free?

Yes. There's no trial, no word limit and no paid plan. SpeakType is open source under the MIT license, so anyone can read or reuse the code.

Does anything get sent over the internet?

No. Your voice is turned into text on your computer. SpeakType only goes online to download speech models when you ask it to, and to check for new versions, which you can turn off in Settings.

Which model should I pick?

The one SpeakType recommends is a good start. Roughly: smaller models are quicker and use less memory, and larger ones make fewer mistakes. Parakeet models are very fast and cover English plus 24 other European languages. Whisper models cover 99 languages. You can download several and switch any time.

Where are my recordings and history kept?

Only on your computer. You can delete single dictations or clear everything from the History screen.

I used the old Mac app. Do I lose my history?

No. The next SpeakType 2 update brings over your history, dictionary and settings from SpeakType 1 automatically the first time you open it. Until then, SpeakType 1.3 remains available.

Something isn't working

SpeakType 2 is new, and Windows and Linux support is especially young. Please open an issue describing what happened and which computer you're on. It helps a lot.

What's next

• Smarter cleanup, still private: an optional step that tidies up and formats what you said, using an AI model on your computer.
• iPhone and Android: a SpeakType keyboard, with optional syncing between your devices.
• Signed apps for Mac and Windows, so installing takes no extra steps.

Help build SpeakType

Everyone is welcome, and you don't have to write code. Bug reports, ideas and design feedback all help.

The project is new and the code is small and easy to find your way around, so it's a friendly place to learn. People who use Windows or Linux every day are especially welcome.

Start with the contributing guide. It explains how to set up your computer, run and test SpeakType, and send your first change.

---

_{Speech recognition by OpenAI Whisper and NVIDIA Parakeet, running on whisper.cpp, ONNX Runtime, WhisperKit and FluidAudio. MIT licensed.}