§ Raw mode
Raw mode
Raw inserts what speech-to-text produced — nothing more. No dictionary, no replacements, no model rewrite, no network. The shortest path between speaking and seeing text.
Who this is for
Users who want to revise text themselves, journalists pulling quotes, anyone running a transcript through downstream tooling that expects the unmodified STT output.
What Raw includes
Just the model output. WhisperKit (or whichever local STT engine you chose) returns text; Dettivo inserts it. Filler words stay in. False starts stay in. Speaker hesitations stay in.
Punctuation depends entirely on the STT model — most modern models produce reasonable commas and periods. If you want clean punctuation, choose Polish.
Three ways to use Raw
As the default
Settings › AI Processing › Dictation mode → Raw. Every Push-to-Talk session inserts verbatim.
As a one-off hotkey
Bind Insert Raw to a chord you can remember. Default is undefined; common choice is Shift+Option+Space.
In Quick Access
⌘+Return in Quick Access inserts the highlighted transcript as Raw, no matter how it was originally captured.
Launching April 2026
Be first on the launch list.
Dettivo opens for new purchases in April 2026. One-time license, three Macs, lifetime updates.