Dictate text into any Mac app by holding a shortcut key and speaking.
Get an on-device alternative to cloud dictation apps that charge monthly fees.
Clean up spoken transcripts automatically, removing filler words and false starts.
Use an inline voice command to ask a question or rewrite text you just dictated.
| kuberwastaken/megaphone | joeseesun/qmprompter | oomol-lab/lockime | |
|---|---|---|---|
| Stars | 88 | 87 | 89 |
| Language | Swift | Swift | Swift |
| Setup difficulty | easy | moderate | easy |
| Complexity | 2/5 | 3/5 | 1/5 |
| Audience | general | general | general |
Figures from each repo's GitHub metadata at analysis time.
Requires macOS 26 on Apple silicon, unsigned releases trigger a macOS security warning unless the quarantine flag is cleared.
Megaphone is a free, open source dictation app for macOS that runs entirely on your own Mac, with no cloud servers, API keys, or subscriptions involved. You hold the Fn key, speak, and let go, and Megaphone types what you said into whatever app is active. The app is built on two of Apple's newer on-device frameworks. SpeechAnalyzer handles the actual speech to text conversion, and according to a benchmark cited in the README it is both more accurate and roughly three times faster than Whisper Small, while running fully on the Mac itself. Foundation Models, Apple's on-device language model, then cleans up the raw transcript by removing filler words, fixing false starts like saying one word and correcting it to another, and tidying punctuation, all without sending anything off the device. If Apple Intelligence is unavailable or too slow to respond, Megaphone falls back to a simpler, deterministic cleanup pass that just strips obvious ums and stutters without needing a language model at all. Beyond basic dictation, Megaphone includes an alpha inline AI mode you can trigger by starting a dictation with a phrase like Hey Megaphone, which can answer a question or rewrite the text you just spoke, using recent context from the active app. It also supports a personal dictionary for names and technical terms, works with multiple languages that Apple's on-device model supports, and lets you customize shortcuts, sounds, and other behavior. Installation is a signed DMG download for Apple silicon Macs, with a one-line curl command provided for quick setup, though since the app is not notarized with a paid Apple Developer account, macOS will warn about it unless you clear the quarantine flag or build from source yourself. The app requires macOS 26, since SpeechAnalyzer does not exist on earlier versions, so older Macs are not supported. Megaphone began as a fork of an app called FreeFlow, and the author replaced its cloud-based transcription with the on-device Apple stack described above, then added the cleanup, dictionary, and inline AI features on top.
A free macOS dictation app that transcribes speech entirely on-device using Apple's SpeechAnalyzer and Foundation Models, with no cloud or subscription.
Mainly Swift. The stack also includes Swift, macOS, SpeechAnalyzer.
Free and open source, check the repository's license file for exact reuse terms.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly general.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.