Arabic Word and Character Counter

Count words, characters and sentences in Arabic text, with tashkeel and tatweel kept out of the letter count.

How to use

What it checks and how it works

Words and sentences come from Intl.Segmenter with the Arabic locale, which follows the Unicode text segmentation rules (UAX #29), so punctuation and spaces are not counted as words. Characters are reported three ways: code points, visible characters (grapheme clusters, where a letter and its tashkeel count as one) and Arabic letters only. Tashkeel marks and tatweel are counted on their own. A word counter written for English often counts Arabic text wrongly, because it splits on spaces only and counts every mark as a character. The counts here depend on your browser's Unicode data, so the same text can differ by a word between browsers in rare cases.

Why do other counters show a different character count?

Counters differ on whether tashkeel, tatweel and spaces count. This tool shows code points, visible characters and Arabic letters separately so you can pick the one your limit uses.

How are mixed Arabic and English texts counted?

Each word is classed by its script, so you see Arabic words and Latin words separately, and numbers are counted apart.

Is my text uploaded?

No. The counting runs in your browser and nothing leaves it.

Sources

Related tools

Privacy

Runs in your browser; nothing is uploaded. The interactive tool needs JavaScript.