Dictation for your Mac, and transcription for the long stuff. Hold Right Option (⌥), talk, and AFK inserts the result into the focused editable app or keeps it on your clipboard. One purchase, with no per-minute meter.

Drop supported audio into the app for multi-hour transcription in bounded chunks, with word timings and plain-text, .srt, or .json exports. The headless CLI accepts the same audio inputs plus MP4 files with an audio track, keeps transcript output on stdout, and includes a short Agent Skill for coding agents.

Requirements

• macOS 15 (Sequoia) or later, on Apple Silicon (M1 or newer)
• First launch downloads and prepares an on-device speech model. Allow an internet connection and roughly 2 GB of free disk for setup; audio recognition can run offline after that.
• Microphone is used for dictation and Accessibility for the global shortcut and cross-app insertion. AFK processes audio on your Mac and does not upload it. License activation and update checks are separate network operations.

What you get

• A Developer ID signed and Apple-notarized .dmg with in-app updates
• A license key for up to three Mac activations