Remove Accents from Text

How to use it

1. Paste or type your text in the input box, or click Open file to load a .txt, .csv or similar plain-text file. Click Example to try it with a sample.
2. Leave Also convert ß, æ, ø, ł, œ and similar letters ticked to get plain A–Z output, or untick it to remove only the marks and leave those letters alone.
3. The result updates as you type. Click Copy, or Download .txt to save it as a file.
4. To keep working on the result, click Use as input, or continue in the URL slug generator or the lowercase converter.

Only letters change. Spaces, tabs, line breaks, punctuation, digits and emoji (👍🏽, 🇫🇷, 1️⃣, ❤️) come out as they went in, and the line count stays the same, so a list of 500 names gives back 500 names in the same order.

What the option changes

Most accented letters are a plain letter plus a mark, and removing the mark is enough. A few letters are not built that way: ø is not “o with a mark” in Unicode, and ß and æ are letters in their own right. The option decides what happens to them.

With the option on, the tool converts ø, ß (and capital ẞ), æ, œ, ł, đ, ð, þ, dotless ı and ħ, as listed in the sidebar table, plus a few rarer letters: Catalan ŀ (l), Dutch ij (ij), Sami ŧ (t) and ŋ (ng), long ſ (s), ƒ (f), ĸ (q) and the typographic ligatures ff, fi, fl, ffi, ffl, ſt and st (ff, fi, fl, ffi, ffl, st, st). Upper and lowercase are kept: Ø becomes O and ẞ becomes SS. A capital that turns into two letters (Æ, Œ, Þ, Ŋ) follows the word: Æble becomes Aeble and Þór becomes Thor, but in capitals ÆBLE becomes AEBLE and ÞÓR becomes THOR. IJ is always IJ. Turn it off when you only want to drop the marks and keep the letters a reader of Danish, German or Polish would expect, for example to compare spellings.

How accents are removed

The tool uses Unicode normalization, the same mechanism that lets software treat “ñ” typed as one character and “n” followed by a combining tilde as the same text:

1. The text is converted to NFD (canonical decomposition), which splits each accented letter into its base letter and one or more combining marks. Ç becomes C plus a combining cedilla.
2. The combining marks on Latin and Greek letters are deleted. On Cyrillic letters only stress marks are deleted, and ё becomes е; marks in other scripts are kept.
3. The text is converted back to NFC, so anything left over is stored in its usual composed form.
4. If the option is on, the letters and ligatures listed above are replaced one by one.

Because this works on Unicode data rather than a hand-made list, it covers letters from every Latin-alphabet language, including stacked marks such as Vietnamese ễ and ặ, the double acute in Hungarian ő and ű, and the comma below in Romanian ș and ț.

Some characters are left untouched because they are not accented letters: fractions such as ½, symbols such as ™ and full-width letters such as Ａ. Ligatures such as fi and rare letters such as ŋ, ŀ and ij change only when the option is on. Check the output if your text contains them, and use find and replace for any you want to change.

Worked examples

A list keeps its line breaks and commas:

Müller, Jürgen
García, José
Østergaard, Søren
Wałęsa, Łukasz

becomes:

Muller, Jurgen
Garcia, Jose
Ostergaard, Soren
Walesa, Lukasz

Note that the curly apostrophe in l’hôtel and the en dash in a file name such as Résumé – Ana Peña (2026).pdf (which becomes Resume – Ana Pena (2026).pdf) are not letters, so they stay. The special characters remover deals with those.

When people need it

• Names in spreadsheets and CRMs. Older systems, mail-merge tools and some payment or airline forms reject or garble accented letters. Cleaning a name column first gives Jose Garcia instead of José García.
• File names. Accented file names can break when files are moved between Windows, macOS, Linux and web servers, or when they are zipped and emailed. Plain letters avoid the problem.
• Search and matching. People type cafe and creme brulee without accents. Removing accents from both your data and the search terms lets them match, and the same applies when you look for duplicates in a mailing list.
• Usernames, codes and IDs that only allow A–Z, 0–9 and a few symbols.
• URLs. For web addresses, go straight to the slug generator: it removes accents, lowercases and adds hyphens in one step, so Crème Brûlée & Café Olé becomes creme-brulee-and-cafe-ole. The guide what is a URL slug explains why plain ASCII slugs are easier to share.

When not to remove accents

In published text, accents change meaning and pronunciation: Spanish año (year) and ano are different words, as are French côte and cote. Keep accents in anything people will read, and remove them only for systems that cannot handle them.

The tool is designed for the Latin alphabet and also strips Greek accents (Ἀθῆναι becomes Αθηναι). It leaves alone marks that are part of a letter in other scripts: Russian й stays й (only stress marks go, and ё becomes е), Japanese が stays が, and Hindi, Thai, Hebrew and Arabic text comes out unchanged, so mixed-language text is safe to paste. Compare before and after with the text compare tool if you are unsure what changed.

Questions