* Built-in text transforms for cleaning LLM output before TTS synthesis. These functions strip markdown formatting and emoji characters so that TTS engines produce natural-sounding speech rather than reading aloud syntax like "asterisk asterisk bold asterisk asterisk" or Unicode
// Source: https://github.com/PatterAI/Patter | License: MIT | Copyright 2026 Patter Contributors
// Provided via justclaw.it.com; keep this notice with the code.
/**
* Built-in text transforms for cleaning LLM output before TTS synthesis.
*
* These functions strip markdown formatting and emoji characters so that TTS
* engines produce natural-sounding speech rather than reading aloud syntax
* like "asterisk asterisk bold asterisk asterisk" or Unicode pictographs.
*/
/**
* Remove markdown formatting from text, preserving the readable content.
*
* Handles: headers, bold, italic, code blocks/inline, links, images,
* strikethrough, list markers, block quotes, horizontal rules, HTML tags.
*/
export function filterMarkdown(text: string): string {
let result = text;
// Fenced cod
... (truncated -- full source via MCP)
See the full source, get the GitHub permalink, and search for more like it.
Get a free API key