Remove Text Formatting

When you copy text from websites, Word documents, or PDFs, it often comes with invisible formatting: font styles, sizes, colors, line breaks, and special Unicode characters. This can cause problems when pasting into plain text editors, forms, or databases.

Common Problems Solved

• Weird Line Breaks: Text copied from PDFs often has random breaks mid-sentence.
• Smart Quotes: Curly quotes (") that don't work in code or databases.
• Hidden Characters: Zero-width spaces, soft hyphens, or other invisible formatting.
• Accents & Diacritics: Removing accent marks to get plain ASCII letters.

What "formatting" actually means in Unicode

Styled text copied from social media is rarely styled in the way a word processor styles text. There's no bold attribute traveling with it. Instead the letters have been swapped for entirely different Unicode characters that happen to look bold. The bold letter in a fancy bio is a Mathematical Bold Capital, a distinct codepoint reserved for mathematical notation, not the letter A wearing a bold style.

This is why the text behaves oddly once you paste it somewhere real. Search won't match it against ordinary words, screen readers may announce it character by character or skip it, and sorting puts it in unexpected places. Converting those lookalike codepoints back to plain ASCII is what this tool does, and it's the only way to make the text function normally again.

The characters you cannot see

Pasted text also carries invisible passengers. Zero-width spaces and zero-width joiners are used to fake line breaks in bios and to stop platforms from collapsing whitespace. Combining marks stack accents above and below letters to produce glitch or Zalgo effects, and a single visible character can carry dozens of them. Because none of these render as anything, a string that looks like ten characters can easily be several hundred bytes, which is how people unexpectedly hit character limits on text that appears short.

Directional overrides are the ones worth knowing about for safety reasons. These control characters flip the reading order of the text that follows, and they can be used to disguise what a filename or a link actually says. Stripping them is good hygiene for anything you did not author yourself.

Normalization, and why it doesn't fix everything

Unicode defines normalization forms that reconcile characters which should be treated as equivalent. The compatibility forms are the aggressive ones, and they will fold many decorative variants back toward their plain equivalents, turning fullwidth letters and mathematical alphanumerics into ordinary ASCII. They don't catch everything, though. Characters chosen purely because they resemble a Latin letter, such as Cyrillic characters used as lookalikes, are genuinely different letters with no compatibility relationship. So normalization leaves them untouched and a substitution pass is still required.

For other text cleanup tasks, try our Remove Underscores tool to convert snake_case to readable text.