Put your cursor anywhere you write.
Open a text field in your app. Sayble waits quietly in the menu bar.
PRIVATE DICTATION FOR MAC
Hold a key. Say what you mean. Sayble turns your voice into text, right where you’re writing. Your voice and words stay on your Mac.
Apple Silicon · macOS 14+ · 16 GB memory
To Design team
Let’s keep the first version simple. One clear idea, a little room to breathe, and something that feels good to use.
LESS BETWEEN YOU AND YOUR WORDS
For the email, the message, the idea you want to get down before it disappears.
Open a text field in your app. Sayble waits quietly in the menu bar.
Release to finish. For a longer thought, double-tap Fn for hands-free dictation.
Sayble transcribes, applies light cleanup, and inserts the result into the focused field. Review it and carry on.
A PERSONAL TOOL SHOULD STAY PERSONAL
Speech recognition and text cleanup run on your device. Download the speech model once, then dictate without an internet connection.
Read the privacy detailsNo audio uploads.
No cloud transcription.
No account needed for dictation.
FREE EARLY ACCESS
Try the early version and tell us how it works in your day. This test build is free, with no account, card, or license key.
Installation & helpFree during testing
Version 0.1.0 (4) · Apple Silicon
3.9 MB · macOS 14+
Developer ID signed. Notarized by Apple. Early software: expect rough edges.
SHA-256 checksumThe current requirement is an Apple Silicon Mac (M1 or newer), macOS 14 or later, and at least 16 GB of memory. The default speech model downloads about 483 MB; allow additional space for installation and model compilation.
The current speech model supports 25 languages, including English and Russian. Accuracy varies by language, accent, and recording conditions. We do not promise the same quality in every language.
Sayble inserts into the focused text field using macOS Accessibility and clipboard-based insertion. Some custom editors and protected fields may behave differently. Compatibility testing is still in progress.
Microphone access records your speech. Accessibility and Input Monitoring support the global trigger and insertion into other apps. See the setup guide for instructions.
The current version applies light, rule-based cleanup, including selected filler words and repeated fragments. It does not include an AI rewriting assistant, dictation history, or a custom dictionary.