AI-Powered · 120+ Languages

Translate PDF to Kurdish (Sorani)

Convert PDFs into Central Kurdish for readers in Erbil, Sulaymaniyah and western Iran. Right to left Perso-Arabic script, the Kurdish specific letters at their own code points, tables mirrored rather than copied. Files up to 1 GB.

Max. file size 1 GB Keeps original formatting
Sign Up Free

Upload or drop document to translate

Max. file size 1 GB

.PDF .DOCX .PPTX .XLSX .TXT .JPG .PNG .IDML .EPUB .HTML
Afrikaans (Afrikaans)
Shqip (Albanian)
አማርኛ (Amharic)
العربية (Arabic)
Հայերեն (Armenian)
Azərbaycan dili (Azerbaijan)
Euskara (Basque)
Беларуская (Belarusian)
বাংলা (Bengali)
Bosanski (Bosnian)
Български (Bulgarian)
မြန်မာဘာသာ (Burmese)
Català (Catalan)
Cebuano (Cebuano)
Chichewa (Chichewa)
中文 简体 (Chinese Simplified)
中文 繁體 (Chinese Traditional)
Corsu (Corsican)
Hrvatski (Croatian)
Čeština (Czech)
Dansk (Danish)
Nederlands (Dutch)
English (English)
Esperanto (Esperanto)
Eesti (Estonian)
Suomi (Finnish)
Français (French)
Frysk (Frisian)
Galego (Galician)
ქართული (Georgian)
Deutsch (German)
Ελληνικά (Greek)
ગુજરાતી (Gujarati)
Kreyòl Ayisyen (Haitian)
Hausa (Hausa)
ʻŌlelo Hawaiʻi (Hawaiian)
עברית (Hebrew)
हिंदी (Hindi)
Hmoob (Hmong)
Magyar (Hungarian)
Íslenska (Icelandic)
Igbo (Igbo)
Bahasa Indonesia (Indonesian)
Gaeilge (Irish)
Italiano (Italian)
日本語 (Japanese)
Basa Jawa (Javanese)
ಕನ್ನಡ (Kannada)
Қазақ тілі (Kazakh)
ខ្មែរ (Khmer)
Ikinyarwanda (Kinyarwanda)
한국어 (Korean)
Kurdî (Kurdish)
Кыргызча (Kyrgyz)
ລາວ (Laotian)
Latina (Latin)
Latviešu (Latvian)
Lietuvių (Lithuanian)
Lëtzebuergesch (Luxemb)
Македонски (Macedonian)
Malagasy (Malagasy)
Bahasa Melayu (Malay)
മലയാളം (Malayalam)
Malti (Maltese)
Te Reo Māori (Maori)
मराठी (Marathi)
Монгол хэл (Mongolian)
नेपाली (Nepali)
Norsk (Norwegian)
ଓଡ଼ିଆ (Odia)
فارسی (Persian)
Polski (Polish)
Português (Portuguese)
ਪੰਜਾਬੀ (Punjabi)
Română (Romanian)
Русский (Russian)
Gagana Samoa (Samoan)
Gàidhlig (Scottish)
Српски (Serbian)
Sesotho (Sesotho)
Shona (Shona)
سنڌي (Sindhi)
සිංහල (Sinhala)
Slovenčina (Slovakian)
Slovenščina (Slovenian)
Soomaali (Somali)
Español (Spanish)
Basa Sunda (Sundanese)
Kiswahili (Swahili)
Svenska (Swedish)
Tagalog (Tagalog)
Тоҷикӣ (Tajik)
தமிழ் (Tamil)
Татарча (Tatar)
తెలుగు (Telugu)
ไทย (Thai)
Türkçe (Turkish)
Türkmençe (Turkmen)
Українська (Ukrainian)
اردو (Urdu)
ئۇيغۇرچە (Uyghur)
O'zbekcha (Uzbek)
Tiếng Việt (Vietnamese)
Cymraeg (Welsh)
isiXhosa (Xhosa)
ייִדיש (Yiddish)
Yorùbá (Yoruba)
isiZulu (Zulu)
Afrikaans (Afrikaans)
Shqip (Albanian)
አማርኛ (Amharic)
العربية (Arabic)
Հայերեն (Armenian)
Azərbaycan dili (Azerbaijan)
Euskara (Basque)
Беларуская (Belarusian)
বাংলা (Bengali)
Bosanski (Bosnian)
Български (Bulgarian)
မြန်မာဘာသာ (Burmese)
Català (Catalan)
Cebuano (Cebuano)
Chichewa (Chichewa)
中文 简体 (Chinese Simplified)
中文 繁體 (Chinese Traditional)
Corsu (Corsican)
Hrvatski (Croatian)
Čeština (Czech)
Dansk (Danish)
Nederlands (Dutch)
English (English)
Esperanto (Esperanto)
Eesti (Estonian)
Suomi (Finnish)
Français (French)
Frysk (Frisian)
Galego (Galician)
ქართული (Georgian)
Deutsch (German)
Ελληνικά (Greek)
ગુજરાતી (Gujarati)
Kreyòl Ayisyen (Haitian)
Hausa (Hausa)
ʻŌlelo Hawaiʻi (Hawaiian)
עברית (Hebrew)
हिंदी (Hindi)
Hmoob (Hmong)
Magyar (Hungarian)
Íslenska (Icelandic)
Igbo (Igbo)
Bahasa Indonesia (Indonesian)
Gaeilge (Irish)
Italiano (Italian)
日本語 (Japanese)
Basa Jawa (Javanese)
ಕನ್ನಡ (Kannada)
Қазақ тілі (Kazakh)
ខ្មែរ (Khmer)
Ikinyarwanda (Kinyarwanda)
한국어 (Korean)
Kurdî (Kurdish)
Кыргызча (Kyrgyz)
ລາວ (Laotian)
Latina (Latin)
Latviešu (Latvian)
Lietuvių (Lithuanian)
Lëtzebuergesch (Luxemb)
Македонски (Macedonian)
Malagasy (Malagasy)
Bahasa Melayu (Malay)
മലയാളം (Malayalam)
Malti (Maltese)
Te Reo Māori (Maori)
मराठी (Marathi)
Монгол хэл (Mongolian)
नेपाली (Nepali)
Norsk (Norwegian)
ଓଡ଼ିଆ (Odia)
فارسی (Persian)
Polski (Polish)
Português (Portuguese)
ਪੰਜਾਬੀ (Punjabi)
Română (Romanian)
Русский (Russian)
Gagana Samoa (Samoan)
Gàidhlig (Scottish)
Српски (Serbian)
Sesotho (Sesotho)
Shona (Shona)
سنڌي (Sindhi)
සිංහල (Sinhala)
Slovenčina (Slovakian)
Slovenščina (Slovenian)
Soomaali (Somali)
Español (Spanish)
Basa Sunda (Sundanese)
Kiswahili (Swahili)
Svenska (Swedish)
Tagalog (Tagalog)
Тоҷикӣ (Tajik)
தமிழ் (Tamil)
Татарча (Tatar)
తెలుగు (Telugu)
ไทย (Thai)
Türkçe (Turkish)
Türkmençe (Turkmen)
Українська (Ukrainian)
اردو (Urdu)
ئۇيغۇرچە (Uyghur)
O'zbekcha (Uzbek)
Tiếng Việt (Vietnamese)
Cymraeg (Welsh)
isiXhosa (Xhosa)
ייִדיש (Yiddish)
Yorùbá (Yoruba)
isiZulu (Zulu)
ARABIC PORTUGUESE RUSSIAN ITALIAN KOREAN DUTCH POLISH TURKISH SWEDISH ENGLISH SPANISH FRENCH GERMAN CHINESE JAPANESE HINDI BENGALI VIETNAMESE THAI GREEK HEBREW ARABIC PORTUGUESE RUSSIAN ITALIAN KOREAN DUTCH POLISH TURKISH SWEDISH ENGLISH SPANISH FRENCH GERMAN CHINESE JAPANESE HINDI BENGALI VIETNAMESE THAI GREEK HEBREW

An Arabic script that writes all of its vowels

Central Kurdish, which its speakers call Sorani, is written in a Perso-Arabic alphabet and runs right to left. The alphabet is adapted rather than borrowed. It adds letters Arabic does not have, among them ڕ for a rolled r, ڵ for a dark l, ۆ for o, ێ for a long e and ڤ for v, and it takes چ, ژ, پ and گ from Persian. What sets it apart from most Arabic derived scripts is that it is fully alphabetic. Arabic and Persian leave short vowels unwritten and rely on the reader to supply them, while Sorani spells every vowel out as a letter. That makes the script easier for a learner and harder for a font, because the letters carrying those vowels are exactly the ones a general purpose Arabic typeface is most likely to lack.

One letter deserves its own warning. The character ە carries the short a sound, it is one of the most frequent letters in the language, and it is a different code point from the Arabic letter ه that it closely resembles. Text typed on an Arabic keyboard, or run through a process that normalises Kurdish towards Arabic, ends up holding the wrong character. The page then looks almost right and is subtly wrong from beginning to end, which is far more damaging than an obvious failure because nobody notices in time. The same trap catches ی and ك, where Kurdish uses the Persian forms and Arabic input produces the Arabic ones.

The script is cursive. Each letter takes one of up to four shapes depending on whether it stands alone, opens a word, sits inside one or closes it, and the choice is made by the font at layout time rather than stored in the file. A font missing the Kurdish letters therefore cannot join them either, so those letters drop out of the connected line and float in isolation. A Kurdish reader spots that instantly, even without being able to name what has gone wrong.

What a right to left page demands that a left to right one does not

Mixed direction on almost every line

Product names, email addresses, model numbers and Western digits sit inside Kurdish sentences constantly. Unicode defines an algorithm for laying out those mixed runs, and output that ignores it puts the pieces of a line in an order that reads as scrambled even though every character is correct.

Brackets and quotes mirror

An opening parenthesis in a right to left run is drawn as the mirror image of the same character in a left to right one. Where a document is full of clause references, citations and units, getting the mirroring wrong turns ordinary punctuation into visual noise.

Tables have to flip, not just translate

Column order runs right to left, headers align to the right and the reading path across a row reverses. A table whose cells have been translated but whose geometry was copied from the English original reads backwards to the person it was made for.

Three sets of digits are in play

Kurdish material from Iraq often uses Arabic-Indic numerals, material from Iran uses the Persian variants, and technical and financial text frequently uses Western ones. Documents mixing two sets are common, so the numeral style is a decision to make rather than something to leave to a default.

Sorani or Kurmanji, and why the script settles it

Kurdish is not one written language. Sorani is spoken across the Kurdistan Region of Iraq, around Erbil, Sulaymaniyah and Halabja, and in the neighbouring provinces of western Iran, and it is written in the Perso-Arabic alphabet described above. Northern Kurdish, called Kurmanji, is spoken by the larger number of Kurds in Turkey and Syria and in parts of northern Iraq, and it is written in a Latin alphabet devised in the 1930s. The two are relatives, not variants: they differ in grammar as well as in spelling, and a Kurmanji speaker in Diyarbakır cannot simply read a Sorani document. There is no automatic conversion between them, so choosing the wrong one produces a file nobody in the intended audience can use.

The practical rule follows the administration. Kurdish is one of the two official languages of Iraq alongside Arabic, and Sorani is the working language of the Kurdistan Regional Government, so documents issued in Erbil or Sulaymaniyah are Sorani. Material intended for Kurdish readers in Turkey, Syria or the European diaspora that came from those countries is Kurmanji. If you are unsure which you are holding, look at the script: Arabic letters mean Sorani, Latin letters with cedillas and circumflexes mean Kurmanji.

Documents that move between English and Sorani

Two forces drive most of the demand: a large and long established diaspora in western Europe, and an energy sector in northern Iraq that operates in English while filing in Kurdish and Arabic. The usual material is:

  • Iraqi civil documents issued in the Kurdistan Region: identity cards, nationality certificates, and birth, marriage and death records
  • Asylum and protection files for Kurdish applicants in Germany, Sweden, the United Kingdom and the Netherlands, often resting on documents decades old and poorly scanned
  • Transcripts and degree certificates from Salahaddin University in Erbil and the University of Sulaimani
  • Petroleum sector agreements, service contracts and regulatory correspondence from the Kurdistan Region's oil and gas industry
  • Court judgments, property deeds and inheritance settlements
  • Medical records and treatment summaries, frequently prepared for patients being referred to hospitals abroad
  • Humanitarian assessments and programme material for displaced populations across northern Iraq

Check the language on the page before you start: many documents issued in the Kurdistan Region are in Arabic rather than Kurdish, and some carry both. For anything filed with an immigration authority or a court, a certified translation from a qualified human translator is required, and asylum files in particular are read closely. See legal translation for contract and court work.

Sorani Kurdish PDF translation pricing

Start with the 7-day trial and upgrade as your translation needs grow.

7-Day Trial

MOST POPULAR
$2.00 today

then $14.99/month after trial ends

  • 7-day full access trial
  • Trial limit: 10 pages or 3,000 words
  • $0.005/word AI translation
  • 120+ languages
  • PDF, DOCX, XLSX, PPTX, IDML, TXT, JPG, PNG, CSV, JSON
  • Team access & custom glossaries
  • Email support

Monthly

POPULAR
$14.99/month

Regular price $29.99, now 50% off

  • 100 pages or 30,000 words per month
  • $0.005/word AI translation
  • 120+ languages
  • Unlimited file storage
  • PDF, DOCX, XLSX, PPTX, IDML, TXT, JPG, PNG, CSV, JSON
  • Team access & custom glossaries
  • Priority email support
🎉 Best value: save $44.88/year

Annual

SAVE 25%
$135/year

~$11.25/month, save 25% vs monthly

  • 100 pages or 30,000 words per month
  • $0.005/word AI translation
  • 120+ languages
  • Unlimited file storage
  • PDF, DOCX, XLSX, PPTX, IDML, TXT, JPG, PNG, CSV, JSON
  • Team access & custom glossaries
  • Priority email support
Steps required

How to translate your PDF into Sorani Kurdish

01

Create a free account

Sign up with your email to access the online translation dashboard.

02

Upload your PDF file

Drag and drop the file or browse for it. Older Iraqi civil documents are usually scans, and image PDFs are supported alongside digital ones.

03

Choose Kurdish (Sorani) as target language

Set the source language, then pick Sorani. Output is right to left Perso-Arabic script with ڕ, ڵ, ۆ, ێ and ە written at their own code points rather than as Arabic lookalikes.

04

Translate and download

Run the job and open the result. Check one table and one line containing Latin text or digits: those are where a broken bidirectional layout shows up first.

Sorani Kurdish PDF translation FAQ

Sorani or Kurmanji: how do I know which one I need?

Look at the script and at the audience. Sorani is written in Perso-Arabic letters and is the language of the Kurdistan Region of Iraq and of Kurdish areas in western Iran. Kurmanji is written in a Latin alphabet and covers Kurdish readers in Turkey and Syria and much of the western European diaspora that came from there. They differ in grammar as well as in spelling, so one is not a fallback for the other, and there is no automatic conversion between the two.

Will the Kurdish specific letters render, or will I get boxes?

Output embeds a font that carries the full Kurdish set, including ڕ, ڵ, ۆ, ێ and ڤ, which many general purpose Arabic fonts do not have. The failure to watch for is subtler than a hollow box: a font that lacks a letter also cannot join it to its neighbours, so the missing letters break out of the cursive line and sit apart. If a word looks like it has been cut in half, that is what has happened.

What is the problem with the letter ە?

It carries the short a sound, it is one of the most common letters in Sorani, and it occupies a different code point from the Arabic letter it resembles. Text entered on an Arabic keyboard, or passed through software that normalises Kurdish towards Arabic, quietly picks up the Arabic character instead. Nothing looks broken, searches stop matching, spell checking goes wrong, and the document is incorrect throughout. Output should use the Kurdish character, and a source that already has the wrong one is worth cleaning before translating.

Do my tables survive the switch to right to left?

They are rebuilt rather than copied. In a Sorani document the column order runs right to left, headers align right and a row is read from the right hand edge inwards. Translating the cell contents while leaving the geometry as it was in English produces a table that a Kurdish reader has to work backwards through. The same applies to numbered lists, headers and footers, and anything with a left aligned label and a right aligned value.

Which numerals will the translation use?

That depends on where the document is going and it is worth stating up front. Kurdish material produced in Iraq commonly uses Arabic-Indic numerals, material from Iran uses the Persian forms of the same digits, and technical, financial and scientific text frequently keeps Western numerals throughout. Mixed documents are normal rather than exceptional, so the safest approach is to decide the convention before translating rather than after.

My document from the Kurdistan Region is in Arabic, not Kurdish. What now?

That is very common. Iraq issues a great deal of official paperwork in Arabic even within the Kurdistan Region, and some documents carry Arabic and Kurdish side by side. Identify the language actually printed on the page before choosing a source language, because selecting Kurdish for an Arabic document will not produce a usable result. If both languages appear, decide which block the receiving authority needs.

Is an AI translation good enough for an asylum file?

For understanding a document and preparing a submission, yes. For the submission itself, no. Asylum and protection files are read closely, the documents supporting them are often old and poorly scanned, and a mistranslated date or place name creates a credibility problem that is hard to undo later. Use the machine output to work out what you are holding, then have a qualified translator produce the certified translation that gets filed.

Translate your PDF into Sorani Kurdish

DocTranslator converts PDFs into Central Kurdish online, writing the Kurdish letters at their own code points, laying out mixed direction lines correctly and mirroring tables rather than copying them, on files up to 1 GB.

Our Partners

Accenture
Bloomberg
Citrix
P&G
SAP