Text Replacements

Text replacements automatically transform your input text before it's sent to the TTS engine. This is useful for expanding abbreviations, correcting common typos, or substituting words.

How It Works

You type text in the overlay
Before TTS synthesis, the TextReplacementService applies all enabled rules
The transformed text is sent to the TTS engine
The original text in the overlay is unchanged

Managing Rules

In Settings → Replacements:

Replacements tab showing a list of text replacement rules with trigger/replacement fields and enable toggles.

Rule Fields

Field	Description
Trigger Text	The text to find
Replacement Text	The text to replace with
Enabled	Toggle the rule on/off
Case Sensitive	Match exact case (default: off)
Whole Word Only	Only match whole words (default: off)

Operations

Add — Create a new replacement rule
Delete — Remove a selected rule
Import — Import rules from a JSON file
Export — Export rules to a JSON file

Rule Processing

Rules are processed in a single pass:

All enabled rules are sorted by their internal sort order
Each rule finds all matches in the input text
Earlier rules claim their match positions first
Later rules skip any text that overlaps with already-claimed positions
The output is built by walking through all substitutions in position order

Priority

Add important rules with lower sort order numbers so they take priority over overlapping matches from later rules.

Examples

Trigger	Replacement	Case Sensitive	Whole Word	Input → Output
`brb`	`be right back`	No	Yes	"brb" → "be right back"
`lol`	`laughing out loud`	No	Yes	"lol" → "laughing out loud"
`API`	`A-P-I`	Yes	No	"API key" → "A-P-I key"
`btw`	`by the way`	No	Yes	"btw, I'm here" → "by the way, I'm here"

Global Toggle

The entire text replacement system can be enabled or disabled with the Enable Replacements toggle at the top of the Replacements tab. When disabled, no rules are applied regardless of individual rule states.