What does the Urdu typing tool do?
The Urdu typing tool turns Roman letters into Urdu script as you type. You write in the alphabet your keyboard has and the output box fills with Urdu alongside it, which means Urdu text can be produced on a machine with no Urdu layout installed and on a keyboard with no Urdu keycaps. It is a phonetic aid rather than a transliteration standard: letters are matched by sound, longer combinations are matched before shorter ones so that kh, gh, ch and sh each produce a single letter, and punctuation is converted to its Urdu equivalent.
The mapping is worth understanding because it explains the output. Doubling a consonant unlocks the retroflex and aspirated forms — tt gives ٹ, th gives تھ, dd gives ڈ — and vowel pairs distinguish long from short, so aa gives آ while a alone gives ا. An option switches digits to Urdu numerals, so 123 becomes ۱۲۳. Text is set right-to-left in the output box, because that is how Urdu reads, and the file can be copied or downloaded as a plain text file.
There is one honest limitation stated up front. Urdu has more consonant letters than English has consonant sounds, so several distinct Urdu letters share a single Roman spelling: the sound written s can be س, ص or ث, and z can be ز, ذ, ض or ظ. A phonetic scheme has to pick one, and only the word itself or a dictionary knows which belongs. The output is therefore close enough to read and edit, and always wants a proofread before it is sent or published.
How to type in Urdu without an Urdu keyboard
- Type into the left box in Roman letters, using ordinary English spelling of the sounds.
- Watch the right box fill with Urdu script as you go; the two stay side by side for comparison.
- Type doubled consonants for retroflex and aspirated letters — tt, dd, th, dh, ch, kh and sh.
- Use aa for a long a sound; a single a gives the short form.
- Tick the numeral option if you want digits written as ۱۲۳ rather than 123.
- Read the Urdu through, correct any letter the mapping chose wrongly, then copy it or download it as a text file.
Urdu Typing Tool limits and rules
| Item | Limit |
|---|---|
| Direction | Roman input left to right, Urdu output right to left |
| Mapping | A phonetic table, with longer key combinations matched first |
| Digraphs | kh, gh, ch, chh, sh, zh, ph, bh, th, dh, jh, ng and rh are each one letter |
| Retroflex | Doubled letters: tt gives ٹ, dd gives ڈ, rr gives ڑ, nn gives ں |
| Punctuation | A comma becomes ، and a question mark becomes ؟ |
| Numerals | Optional Urdu digits, ۰ to ۹, in place of Western ones |
| Not covered | Translation, a spell checker, correct Urdu orthography and the choice between letters that share a sound |
When to use it
- Writing an Urdu message on a laptop or phone with no Urdu layout installed.
- Drafting Urdu text for a form, a caption or a comment before polishing it by hand.
- Producing Urdu script without learning a keyboard layout or hunting characters in a chart.
- Typing a greeting or a short note where installing a layout would be more trouble than it is worth.
- Converting a Roman Urdu draft into Urdu script to check how it reads.
Is it private?
The conversion happens as you type, in this page, using a table built into it. Nothing is uploaded, stored or logged, and the text disappears when the tab is closed.
Frequently asked questions
No, and the distinction is important. It does not know what any word means; it maps letters to letters by sound. So typing an English word produces Urdu letters arranged to sound like that word, not the Urdu word for the thing. Roman Urdu — Urdu written in Latin letters, as it commonly is in messages — converts sensibly, because the sounds are already Urdu sounds. English prose does not, because it is a different language that happens to share an alphabet.
Because Urdu has more letters than the sound needs, and this tool commits to one for each Roman spelling. The s sound can be written with three different letters depending on the word's origin, and z with four; a phonetic table cannot tell them apart because the distinction is not audible. Only the word's spelling — and so a dictionary — reveals which is correct. That is why the output is a draft to be corrected rather than a final text.
Almost certainly it is a font problem rather than a text problem. Urdu is written in a cascading script and needs a Nastaliq or Naskh font to render properly; without one the browser falls back to a font that shows the letters separated or flattened, which looks broken but copies and pastes correctly. The output box asks for a Nastaliq font by name, so a device with one installed — most phones and many desktops in the region — will show it correctly.
Because Urdu does not normally write short vowels. The diacritics that mark them exist but are optional in ordinary text, and leaving them out is standard practice. A scheme that emitted a letter for every Latin vowel would produce something that looks over-specified and foreign to a reader. This tool emits letters for the long vowels, which usually are written, and leaves short vowels to the reader — which is what a native keyboard does too.
Often, yes. Roman Urdu is widely used in messaging, social posts and informal writing, and it is perfectly legible to its readers. The reason to convert to script is context: a formal letter, a document, a form that expects Urdu, or a reader who prefers the script. There is no rule that Roman Urdu is wrong, and plenty of people choose it deliberately. Convert when the setting calls for the script, not because one form is more correct than the other.
Because the surrounding document's direction may not match the text. Urdu is written right to left, so a paragraph containing it should be marked as such; pasted into a field that assumes left-to-right text, the punctuation and any embedded numbers can appear on the wrong side even though the letters are in the right order internally. The characters themselves are correct — the fix is to set the destination's direction to right to left, or to use the downloaded text file, which carries the content unchanged.
Related tools
Written and maintained by Md. Nur Hossain Repon.