Remove Accents Tool

How it is calculated

The tool works in two passes on the text you paste.

Pass 1: a fixed lookup table replaces Latin letters with diacritics by their base letter (é → e, ü → u, ñ → n, ç → c, ş → s, ğ → g, ı → i, İ → I, ø → o, ł → l, đ → d). A few letters expand to several: ß → ss, æ → ae, œ → oe, þ → th, and the capitals AE, OE.

Pass 2: the result is decomposed (Unicode NFD) and any remaining combining marks are deleted, which also catches accents typed as a separate mark.

Letters without an accent, and characters outside the Latin family such as Cyrillic or Chinese, are left exactly as they were, so the output is not always pure ASCII. Upper and lower case are kept. The "changed" figure counts the original characters that were altered, one each, so ß → ss counts as 1 and a text with nothing to strip shows 0. An empty box shows no result.

Common mistakes

• Expecting a pure ASCII result from any text. Cyrillic, Chinese, Arabic and similar scripts pass through unchanged (Привет café becomes Привет cafe), and so do letters outside the table, such as ẞ.
• Using the output as a reversible spelling. The step loses information: Müller and Muller become identical, so you cannot restore the original.
• Assuming German ü becomes ue. The tool gives plain u (Müller → Muller), and ß becomes ss; that is not the usual German transliteration for umlauts.
• Reading the "changed" figure as the length difference. It counts altered characters: Straße Æther Œuvre reports 3, although the text grows by three letters.
• Using the result as a finished URL slug. Capitals, spaces and punctuation are kept, so lower-casing and replacing spaces with hyphens is still your job.