How does it handle compound words?

Urdu compound words like کتابخانہ (library) are written with a zero-width non-joiner between the parts so the ligatures stay separate without a visible space. The counter removes that joiner before splitting, so the compound counts as one word, not two.

What separates words?

Words are split on whitespace and on Urdu and ASCII punctuation such as the full stop ۔, comma ،, question mark ؟, and brackets. Consecutive separators do not create empty words.

Does it count sentences too?

Yes. Sentences are detected by the Urdu full stop ۔ and the question and exclamation marks, giving a rough sentence count alongside the word total.

Why does my count differ from a Latin-script counter?

Generic counters split on every space and would break a ZWNJ-joined compound into pieces, or fail to recognise Urdu punctuation. This tool is tuned to Nastaliq conventions, so the count matches how a reader perceives the words.

Yes. The counting runs entirely in your browser. No text is sent to a server, stored, or logged.

Urdu Word Counter — Gera Tools

Email me this result

Get this tool's output sent to your inbox, plus one useful tool a week. No spam, unsubscribe any time.

Counting Urdu words with a generic word counter gives wrong results because of one quirk: compound words are stitched together with an invisible joiner rather than a space. This counter understands that convention and counts the way a reader actually sees the words.

How it works

Before counting, the text has every zero-width non-joiner (U+200C) stripped, so a compound such as کتاب‌خانہ collapses to a single token instead of splitting at the joiner. The cleaned text is then split on a class of separators:

whitespace  +  ۔ ، ؛ ؟  +  . , ; : ! ? ( )  +  quotes

Empty tokens from consecutive separators are discarded. Sentences are counted by splitting on the Urdu full stop ۔ plus question and exclamation marks. Character counts are reported with and without whitespace.

Example and notes

The line یہ ایک کتاب‌خانہ ہے۔ آپ کیسے ہیں؟ counts as five words and two sentences — the ZWNJ inside کتاب‌خانہ does not inflate the total. Note that the zero-width non-joiner is invisible, so two pieces of text that look identical can have different naive word counts; this tool normalises that difference away.