Cleans the sentence
Removes supported fillers and repairs punctuation, capitalization, spelling, and formatting. Optional on-device cleanup can also tighten false starts and repetition on supported Macs.
LockedIn Flow for Mac
Hold Fn, speak naturally, and polished text appears at the cursor—without sending your voice to a transcription server.
Available now for Apple Silicon Macs on macOS 15 or later. Windows is not released; join the waitlist.
Pause when you need to. Speak in full thoughts or short bursts. LockedIn Flow keeps the meaning and removes the friction.
Removes supported fillers and repairs punctuation, capitalization, spelling, and formatting. Optional on-device cleanup can also tighten false starts and repetition on supported Macs.
Uses Apple’s Voice Focus path on compatible microphones and automatically returns to stable standard capture when needed.
Personal vocabulary remembers names, terminology, and corrections locally, so the words that matter stay right.
Profiles adapt tone and formatting for Mail, chat, notes, and editors without making you switch tools first.
Developer mode turns spoken casing, symbols, braces, and punctuation into code that lands where the cursor is.
Encrypted history and recovery keep completed text available if an app cannot accept the insertion safely.
Recognition, cleanup, vocabulary, and history live on your Mac. Network access is used to download the speech model and check for updates—not to transcribe your words.
Read the security architectureNo transcription server in the middle.
Try every Pro feature available on your Mac for 14 days with no card. Keep the free tier afterward, or unlock unlimited dictation for one person on two Macs.
Every v1.x update included. Fourteen-day refund window.
Dictation audio and text are processed on-device. Network access is used for the one-time speech-model download and optional update checks—not to transcribe your speech.
Pro is $19 once for one person on two Macs, with every v1.x update included. The free tier remains available after the 14-day Pro trial.
An Apple Silicon Mac running macOS 15 or later, plus a one-time speech-model download. Optional generative features require macOS 26 and Apple Intelligence.