Custom Words and Names in Dictation Apps

Build the list from your own corrections

The best source for a custom word list is the editing you already did. Guessing produces entries that never fire and misses the surname you fix three times a week.

1. Open your sent mail and Slack messages from the past two weeks.
2. Note every term you remember retyping or fixing.
3. Add the recurring ones, and stop at about twenty-five.
4. Keep the raw list in a note so you can see what you added and when.

Two weeks of real messages is enough. Anything that matters shows up more than once in that window, and anything that appears once is not worth an entry.

Sort the list by how often you say the term rather than how wrong the transcript looked. A name you say daily that is slightly off costs more time than a spectacular failure you hit once a month.

What earns a slot

Custom vocabulary works because it raises the odds for words the model already considers unlikely. That means specific terms, not general ones.

Add these

• Names of people you write to or about weekly.
• Client, company, and team names.
• Product names, SKUs, repository names, internal tools.
• Acronyms and initialisms from your field.
• Place names and street names you type often.

Skip these

• Common words the engine already gets right.
• One-off names from a single meeting.
• Terms you type but never say out loud.
• Words you can paste from a source of record anyway.

Long identifiers are usually a poor fit. Account numbers and order codes are faster to paste, and asking a speech engine to hear sixteen digits correctly is a losing bet.

Spelling, pronunciation, and how you actually say it

An entry works when it matches what leaves your mouth. Most disappointing lists fail here rather than in the feature itself.

Casing and hyphens are worth being fussy about, since the whole point is not fixing them later. Enter the form you would ship in an email to a client.

Where the app supports variants or aliases, add the pronunciation you use when you are speaking quickly. The careful version you say while testing is not the version you will use at five o’clock.

Where custom words live

Support differs by app and sometimes by engine inside one app, which is why a list that works in one place does nothing in another.

• Auri Voice: custom words are supported on Parakeet. Keep Parakeet as your default when names drive most of your edits.
• Apple Dictation: personal vocabulary support is limited compared with dedicated apps.
• iOS Text Replacement: expands typed shortcuts, so it helps typing rather than recognition.
• Other Mac apps: check whether vocabulary applies to every engine or only some.

If you switch engines to chase accuracy on noise or an accent, check what happens to your list. Losing custom words while gaining a small general improvement is usually a bad trade for anyone who writes about named people.

On a phone, do the same audit before you rely on it. Auri Keyboard offers local speech to text with cloud as the default on iPhone, and vocabulary behavior is worth testing with the names you use rather than assuming.

Test in a way that proves it worked

Adding entries feels productive, so it is easy to skip verification and keep a list full of things that do nothing. Ten minutes of testing fixes that.

1. Write one sentence that uses five of your new terms naturally.
2. Dictate it before adding the entries and count the errors.
3. Add the entries, dictate the same sentence, and count again.
4. Dictate one real email and one real Slack update to confirm it holds.
5. Remove entries that never fired.

Test at your normal speaking pace. Terms pronounced carefully during a test and quickly in real work behave differently, and the real work is what you are optimizing.

Keep the count. Going from six corrections to one on the same sentence is the kind of evidence that tells you to stop tuning and get back to writing.

Maintenance that takes two minutes a month

Vocabulary decays. Clients end, projects close, and the names you needed in March become noise by September.

• Add three to five terms when a project starts.
• Remove terms from finished work once a quarter.
• Re-check the list when you change engines or reinstall.
• Share the list with one teammate so it is not stuck in your head.
• Note new spellings when someone changes their name.

The name change case is the one that stings. Getting a colleague’s new surname wrong in writing because your word list still holds the old one is worse than a generic transcription error.

Sharing the list also makes onboarding trivial. A new teammate who inherits twenty correct spellings for your product and clients skips the two weeks of corrections you went through.

When vocabulary is the wrong fix

A word list solves a narrow problem. Reaching for it when something else is broken wastes an afternoon.

That last row causes real confusion. Auri Voice improves text on device by default, so punctuation, filler removal, and light formatting are expected behavior rather than recognition drift.

When a name keeps splitting into two words no matter what you enter, try a shorter alias you can say cleanly and let the cleanup pass or a quick edit handle the rest. Fighting one stubborn term for an hour is not worth the seconds it costs you.

FAQ