Dettivo

§ Dictation modes

Raw, Polish, Enhanced — pick the right mode

Raw inserts the speech-to-text output verbatim. Polish adds deterministic dictionary + replacement + snippet processing — no model, no network. Enhanced adds context-aware model rewriting on top, optionally translates, and can run automations.

Who this is for

Anyone choosing a default in Settings › AI Processing, or deciding between Insert Raw and Insert Enhanced for a single transcript.

Side by side

StageRawPolishEnhanced
Speech-to-textYesYesYes
Dictionary substitutionsNoYesYes
Replacement rulesNoYesYes
Snippet expansionNoYesYes
Punctuation normalizationNoYesYes
App-target post-processorsNoYesYes
App / window contextNoNoYes
Selected text contextNoNoYes (opt-in)
Window screenshot contextNoNoYes (opt-in)
Model rewriteNoNoYes
TranslationNoNoYes (when enabled)
Macros / automationsNoNoYes (when allowlisted)

How to think about it

Raw is the honest version of what you said. Use it when you want to revise the text yourself, when accuracy matters more than polish, or when you are running a transcription you intend to feed into another tool.

Polish is the deterministic improvement of Raw. It never invents content. It is instant and offline. It is what most people leave on as a default.

Enhanced is for cases where the text needs to be aware of where it lands. Different polish for a Slack reply vs an email vs a code review comment. Optional model rewrite when filler is heavy. Translation if you set a target language. Use Enhanced when context matters more than latency.

Switch mid-session

Hotkeys for Insert Raw and Insert Enhanced override the default for a single transcript. Bind both in Settings › Hotkeys. The menu bar also exposes a Dictation Mode segmented control if you prefer the mouse.

§ Related

Launching April 2026

Be first on the launch list.

Dettivo opens for new purchases in April 2026. One-time license, three Macs, lifetime updates.