Text Cleaner
Clean up and standardize AI-generated text, fix encoding or character issues, and ensure smooth integration for any region or workflow. All processing happens securely in your browser — no text ever leaves your device.
Processing Options
A preset switches the toggles below to a proven combination — fine-tune afterwards if needed.
Character Processing
Remove Invisible Characters & Watermarks
Removes 441 invisible code points: zero-width characters, bidi controls, tag characters and unusual spaces. Catches character-based watermarks and tracking marks. Emoji stay intact.
Apply Character Mapping
Normalizes typography to plain ASCII. Choose which groups below.
Hyphens & dashes
Normalizes every dash and the minus sign to the ASCII hyphen. Off keeps em dashes as they are.
Quotation marks
Normalizes curly quotes, guillemets and the double prime to the ASCII ".
Apostrophes
Normalizes single quotes, primes and spacing accents to the ASCII apostrophe.
Bullets, ellipses & symbols
Ellipsis becomes three dots, bullets become hyphens, × and ⁄ become x and /.
PDF ligatures
PDFs store fi and fl as one glyph. This splits them apart: final → final.
Spell out æ/œ as ae/oe
Turns æ/œ into ae/oe. Leave off for Italian, where they only appear in foreign words.
Convert ß to ss
The Swiss convention. Leave off for German from Germany or Austria.
French spacing rules
French puts a space inside « guillemets » and before ; : ! ? — this removes it.
Remove Diacritics
Turns accented letters into plain ones: é → e, ñ → n.
Umlauts as ae/oe/ue
The German and Swiss convention. Off gives the plain a/o/u.
Remove Fancy Text Font
Normalizes stylized Unicode alphabets — bold, italic, script, fullwidth — back to plain letters.
Content Cleaning
Remove Citation References
Removes markers like [1], superscripts and (Source: ...). Normal links keep their text.
Convert MD to Normal Text
Strips Markdown formatting: headings, bold, italic, lists.
Replace em dash with new sentence
Turns an em dash into a full stop and starts a new sentence.
Remove Line Breaks
Joins everything into one continuous paragraph.