Remove Arabic diacritics (tashkeel)
Strip harakat, tanween, shadda, sukun and Quranic marks from Arabic text, with optional tatweel removal and letter normalisation.
Loading tool…
Fully vowelled Arabic, such as Quranic text, poetry or children's books, carries short-vowel and other marks above and below the letters. They are separate Unicode characters, so vowelled and unvowelled spellings of the same word do not match in search, spreadsheets, databases or spell checkers. This tool removes them and leaves the letters untouched.
By default it removes the combining marks U+064B–U+065F (fathatan, dammatan, kasratan, fatha, damma, kasra, shadda, sukun, and the less common maddah, hamza and subscript marks), the superscript (dagger) alef U+0670, and the Quranic annotation signs U+06D6–U+06ED, as listed in the Unicode Arabic code chart. The counts under the result show how many characters were removed.
Four further options are off by default because they change spelling rather than remove marks: removing tatweel (the kashida stretching character, U+0640), turning أ إ آ into bare alef ا, alef maqsura ى into yaa ي, and ta marbuta ة into haa ه. They are useful for building search indexes that match regardless of how a word was typed, but not for text people will read.
How to use it
- Paste the Arabic text, or click "Try a sample".
- Keep "Remove tashkeel" on, and switch on tatweel removal or letter normalisation only if you need them for searching or matching.
- Copy the result or download it as a .txt file.
Frequently asked questions
Which characters are removed?
U+064B to U+065F (harakat, tanween, shadda, sukun and related marks), U+0670 (superscript alef) and U+06D6 to U+06ED (Quranic annotation marks). Letters, hamza on its own seat (أ ؤ ئ) and punctuation are kept.
Is hamza removed?
Hamza written as part of a letter (أ, إ, ؤ, ئ) is a separate letter in Unicode and stays. Only the combining hamza marks U+0654 and U+0655 are removed with the tashkeel. The optional alef normalisation turns أ, إ and آ into ا.
Why normalise ى to ي or ة to ه?
People type these interchangeably, especially in Egypt and on phones, so search engines often treat them as equal. Normalising both your data and your queries makes matching reliable. It also produces misspellings in the output, which is why it is off by default.
Does it work with Persian or Urdu text?
Yes for the shared marks. Persian and Urdu letters such as پ, چ, گ and ی are letters, not diacritics, and are never changed.
Related tools
- Add Arabic diacritics (tashkeel) automaticallyAdd harakat (fatha, damma, kasra, sukun, shadda, tanween) to Arabic text automatically in your browser, then fix any letter with a manual tashkeel toolbar.
- Convert Arabic numerals to English (Western) numeralsConvert Arabic-Indic digits (٠-٩) and Persian digits (۰-۹) to 0-9 and back, including the Arabic decimal and thousands separators.
- Word counter and character counterCount words, characters (with and without spaces), sentences and paragraphs, estimate reading and speaking time, and see keyword density.
- Find and replace textFind and replace text, as plain text or with regular expressions (capture groups, flags), with a count of replacements.
- URL slug generatorTurn titles into clean URL slugs: accents transliterated, punctuation removed, one slug per line, optional Unicode.
- Remove extra spaces and blank linesTrim lines, collapse repeated spaces, remove or collapse blank lines, convert tabs and spaces, and strip invisible characters.