Using Flunkey

Dictionary & Polish

Add preferred vocabulary and explicit correction rules, then choose the cleanup mode that fits each dictation.

Custom dictionary

Add people, products, acronyms, and domain terms once — “Priya Natarajan”, “Supabase”, “HbA1c”, matter numbers — and Flunkey can prefer your spelling over the model’s guess. You can also define an explicit correction for a known systematic misrecognition, mapping the mistaken transcription to the spelling or phrase you intended. Entries apply longest-first with fuzzy typo repair (Levenshtein, 0.82 confidence), and each entry can be searched, edited, deleted, or toggled on/off.

Entry format

spoken "supabase"      → Supabase
spoken "doctor patel"   → Dr. Patel
spoken "h b a one c"    → HbA1c

Correction rule

recognized "Cloud Flare" → "Cloudflare"
recognized "my product name" → "MyProduct"

Dictionary replacement is skipped in F8 Verbatim mode. In Light and Enhanced modes it can correct recognized text according to your enabled entries.

Grammar polish

F8 Light cleans grammar, punctuation, fillers, stutters, and self-corrections without applying custom prompts. Enhanced applies your global and active-app instructions for more substantial rewriting and formatting. F8 Verbatim skips the second-pass cleanup model and dictionary replacement. F10 has its own model assignment and transform behavior.

Tips that work

Add the spoken form you actually say, not just the correct spelling
Prefer full names (“Priya Natarajan”) over first names for accuracy
Keep entries short — one term per line beats long phrases
Re-test after adding: dictate the tricky sentence once and confirm

Team glossaries (manual for now)

There is no shared cloud dictionary — by design, there is no cloud. Export your list (flunkey-dictionary-YYYY-MM-DD.json) and share the file; import merges on the other side. A one-click import is on the roadmap.

Pill, Sounds & Tray

Shortcuts Reference