Dettivo
Get notified

§ Polish mode

Polish mode

Local-only

Polish takes the Raw transcript and runs a deterministic pipeline on it: dictionary substitutions, replacement rules, snippet expansions, punctuation normalization, code/path/flag post-processing, and insertion formatting. No model call, no network, instant.

Who this is for

Most users, most of the time. Polish is the right default when you want clean output without latency cost or model unpredictability.

What "deterministic" means

Every transformation Polish applies is rule-driven. Your dictionary has fixed entries. Your replacement rules have fixed source → target mappings. Your snippets have fixed triggers. The same Raw transcript will always produce the same Polish output. There is no model temperature, no creative interpretation.

The Polish pipeline

  1. 01

    Dictionary substitutions

    Per-language vocabulary lists and proper-noun grounding. Teaches Dettivo to hear "Dettivo" as Dettivo, not "the TIVO".

    Settings › Dictionary › Vocabulary

  2. 02

    Replacement rules

    Heard → wanted mappings. "Quote unquote" → empty, "h-t-t-p-s colon" → "https:", whatever you want.

    Settings › Dictionary › Replacements

  3. 03

    Snippet expansion

    Trigger words expand to longer strings before insertion. Defined in Settings › General.

    Settings › General › Snippets

  4. 04

    Punctuation + token normalization

    Smart quotes, en/em dashes, ellipsis, consistent capitalization after periods.

  5. 05

    Code / path / flag post-processing

    In editors and code apps, Polish recognizes `/path/to/thing`, `--flag`, and `function()` spans and avoids touching them.

  6. 06

    App-target post-processors

    Per-app rules — different list ordering for Slack vs Mail, code-fence formatting in IDEs.

    Settings › AI Processing › App context

§ 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.