Text Replacements

Text replacements let you automatically transform your transcribed text before it's typed out. Create rules to expand abbreviations, insert symbols, fix common misrecognitions, or perform complex text transformations using regular expressions.

How It Works

After WhisperTyping transcribes your speech, but before the text is typed into your application, all enabled replacement rules are applied in order. This happens before any AI processing, giving you precise control over the raw transcription output.

Processing Order: Speech → Transcription → Built-in Replacements → Your Text Replacements → AI Processing (if enabled) → Output

Creating a Replacement Rule

Open Settings → Replacements and click "Add" to create a new rule. Each rule has the following options:

Aliases (Match Patterns)

Enter one or more text patterns to match. Each alias is checked independently, and any match triggers the replacement. This is useful when:

A word might be transcribed differently (e.g., "email" and "e-mail")
You want multiple phrases to produce the same output
You're creating shortcuts (e.g., "btw" → "by the way")

Replacement Text

The text that will replace any matched alias. Can be:

Plain text
Empty (to delete the matched text)
Multi-line text (for templates or signatures) Pro
Special characters and symbols

Standard Options

These options are available for normal (non-regex) replacement rules:

Enabled Toggle the rule on or off without deleting it

Flexible Match Match aliases regardless of spaces, dashes, colons, semicolons, or periods between characters. "hashtag" will match "hash tag", "hash-tag", "hash.tag", etc.

Absorb Punctuation Include any trailing punctuation (periods, commas, exclamation marks, question marks, semicolons, colons) in the match. Useful when replacing a word that might appear at the end of a sentence.

Preserve Case Maintain the capitalization of the matched text in the replacement. If you match "Hello", the replacement will be capitalized. If you match "HELLO", the replacement will be uppercase.

Scope Options

Control where in the text the replacement can occur:

Anywhere Match the pattern anywhere in the transcription (default)

Whole Word Only match when the pattern is a complete word, not part of another word. "cat" won't match "category"

Start Only Only match at the beginning of the transcription

End Only Only match at the end of the transcription

Common Use Cases

Abbreviation Expansion

Expand shortcuts into full phrases:

Alias	Replacement
btw	by the way
afaik	as far as I know
sig	Best regards, John Smith

Fixing Misrecognitions

Correct words that are frequently mistranscribed:

Alias	Replacement	Tip
whether typing	WhisperTyping	Product names are often misheard
John Doe, john doe	John Doe	Use multiple aliases for name variations

Regular Expressions

For advanced users, enable "Regular Expression" mode to use regex patterns. This gives you powerful pattern matching capabilities.

Important: When regex mode is enabled, Scope, Flexible Match, Absorb Punctuation, and Preserve Case options are hidden and ignored. You have full control through your regex pattern.

Regex Basics

All regex patterns in WhisperTyping are case insensitive by default. You don't need to add any flags for case-insensitive matching.

Pattern	Matches
`hello`	"hello", "Hello", "HELLO", "HeLLo"
`colou?r`	"color" and "colour"
`\d+`	One or more digits
`^hello`	"hello" at the start only
`goodbye$`	"goodbye" at the end only
`\bword\b`	"word" as a whole word

Making Regex Case Sensitive

If you need case-sensitive matching, use the inline modifier (?-i) at the start of your pattern:

Pattern	Matches
`(?-i)Hello`	Only "Hello" (exact case)
`(?-i)API`	Only "API" (uppercase)

Capture Groups in Replacements

Use parentheses to capture parts of the match, then reference them in the replacement with $1, $2, etc.:

Pattern	Replacement	Example
`(\d+) dollars`	`$$$1`	"50 dollars" → "$50"
`(\w+)@(\w+)`	`$1 at $2`	"john@company" → "john at company"

Advanced Regex Examples

\s{2,} → Replace multiple spaces with a single space

^\s+|\s+$ → (empty) Trim leading and trailing whitespace

(\w)'(\w) → $1'$2 Fix curly apostrophes to straight apostrophes

(?<=\d),(?=\d{3}) → (empty) Remove thousands separators from numbers (uses lookbehind/lookahead)

Tip: Insert Your Email Address

Voice transcription can produce different variations of the same phrase. Use .{0,3} (match any 0-3 characters) to catch them all in one rule:

Pattern: work.{0,3}mail → [email protected]

This single pattern matches all of these variations when you say "work email":

"work mail" (space = 1 char)
"work email" (space + e = 2 chars)
"work, mail" (comma + space = 2 chars)
"work, email" (comma + space + e = 3 chars)

Much cleaner than creating multiple separate rules!

Processing Order

Replacement rules are applied in the order they appear in the list. You can drag and drop rules to reorder them. This matters when:

One rule might affect text that another rule matches
You want to apply a broad rule first, then specific exceptions
You're chaining transformations

Tips for Better Replacements

Use Whole Word scope Prevents unintended matches within longer words. "cat" won't match "category"

Test with common variations Add multiple aliases for words that might be transcribed differently

Enable Flexible Match for compound words Catches "email", "e-mail", and "e mail" with a single rule

Use Preserve Case for proper nouns Keeps "HELLO" uppercase and "Hello" capitalized appropriately

Start simple, add regex only when needed Standard options cover most use cases. Use regex for complex patterns

Disable rules instead of deleting You might need them again later

Troubleshooting

Replacement not working?

Check that the rule is enabled
Verify the scope setting matches where the text appears
Try adding the exact transcribed text as an alias (check your history)
If using regex, check for syntax errors in your pattern

Too many matches?

Change scope to Whole Word to avoid partial matches
Add more specific aliases instead of broad patterns
Reorder rules so specific ones come before general ones

Regex not matching?

Remember that patterns are case insensitive by default
Special characters like . * ? + need escaping: \. \* etc.
Test your regex pattern at regex101.com (select .NET flavor)

Built-in Replacements

WhisperTyping comes with many replacements already built in, so you don't need to create them yourself:

Spoken Punctuation – Say "period", "comma", "question mark", "exclamation point", and more to insert punctuation marks
Symbols – Say "at sign", "dollar sign", "hashtag", "ampersand", and other common symbols
Emojis – Say "smiley face", "thumbs up", "heart", and many other emoji names

See the Spoken Punctuation page for a complete list of built-in replacements.

Need help with text replacements?

Having trouble setting up a replacement rule? Send us a message and we'll help you out.