Free tool
Uzbek Latin–Cyrillic converter
Convert text between the two scripts under the official 1995 rules — with every judgement call shown.
- Free
- No ads
- Your text never leaves your device
0 characters
Rules and examples
The converter works from the rules below. Every decision in a result deep-links to its own entry here — which is why this text is the tool’s documentation rather than filler written for search engines.
Latin to Cyrillic
12
A word-initial “e” becomes Cyrillic “э”.
ErtagaЭртага
After a vowel, “e” also becomes “э”.
poemaпоэма
After a consonant, “e” becomes “е” instead.
kelкел
In loanwords the pair “ts” collapses into the single letter “ц”.
sotsialсоциал
Where “t” and “s” meet at a morpheme boundary the result is “тс”, not “ц”.
ketsaкетса
In some Russian loanwords the letter “s” is written “ц”.
sirkцирк
The mark in “oʻ” and “gʻ” is part of the letter itself (U+02BB), not an apostrophe.
oʻzbekўзбек
The tutuq belgisi (U+02BC) is written “ъ” in Cyrillic.
sanʼatсанъат
When “s” and “h” are pronounced separately the result is “сҳ”, not “ш”.
isʼhoqисҳоқ
In Russian loanwords the soft sign dropped by the Latin spelling is restored.
avtomobilавтомобиль
In Russian loanwords the hard sign ъ, which Latin drops, is restored: obyekt → объект.
obyektобъект
The pairs “yo”, “yu”, “ya” fold back into the single letters “ё”, “ю”, “я”.
yozgiёзги
Cyrillic to Latin
18
A word-initial “е” is written “ye”.
ЕрYer
After a vowel, “е” also becomes “ye”.
поездpoyezd
After “ъ” or “ь”, “е” is read as “ye”.
подъездpodʼyezd
An “е” standing at a compound or morpheme boundary is also written “ye”.
After a consonant, “е” is a plain “e”.
келkel
Between two vowels, “ц” is written “ts”.
операцияoperatsiya
A word-initial “ц” is written “s”.
циркsirk
A word-final “ц” is also written “ts”.
шприцshprits
After a consonant, “ц” is written “s”.
концертkonsert
A “ц” on a prefix boundary is decided by word structure rather than by its neighbouring letters.
The cluster “сц” is taken as a unit and becomes “ss”: сценарий → ssenariy.
A doubled “ц” is likewise treated as one cluster rather than two independent letters.
The hard sign “ъ” becomes the tutuq belgisi in Latin.
маъноmaʼno
After “ў”, the hard sign is still the tutuq mark — never the digraph mark.
The soft sign (ь) is not written in the Latin script.
асфальтasfalt
The letters “я” and “ю” after a consonant follow their own rule.
The cluster “сҳ” is not “sh”: the “s” and “h” stay apart.
исҳоқisʼhoq
The letter “щ” is written “shch”.
борщborshch
Capitalisation
3
A word written entirely in capitals stays capitalised after conversion.
OʻZBEKISTONЎЗБЕКИСТОН
A word starting with a capital keeps that capital in the result.
ToshkentТошкент
Inside an acronym both letters of a digraph are capitalised.
AQShАҚШ
Lexicon and direct mappings
2
The word was found in the exception lexicon — proper nouns, place names and loanwords come from there.
A plain one-to-one mapping. Around 90% of any text goes this way and is never listed.
What is left untouched
7
Links (URLs) are never converted.
DOI identifiers are left untouched.
10.1000/xyz12310.1000/xyz123
Email addresses are left alone.
ISSN, ISBN and similar identifiers do not change.
Roman numerals are preserved as written.
XIVXIV
Formulas, chemical symbols and statistical notation are not converted.
Whole runs of English or Russian stay in their own script.
Input repair
3
Incorrect apostrophes were replaced: U+02BB for oʻ/gʻ, U+02BC for the tutuq belgisi.
o'zbekoʻzbek
The text was normalised to Unicode NFC — text pasted from Word or a PDF can spell one letter with two codepoints.
Invisible characters — zero-width spaces and the like — were removed.
Everything else
1
No mapping exists for this character, so it was passed through unchanged.
Using this in academic work
A researcher working on a dissertation, an author’s abstract or a journal article switches scripts more than once: the older sources are in Cyrillic, the current requirements are in Latin, and a supervisor’s comments often arrive with both mixed together. That is why there is no length limit here — one paragraph and a forty-page avtoreferat are handled the same way.
The most damaging mistake in an academic text is a converter that reaches the reference list. If DOIs, ISSNs, email addresses and English-language source titles get transliterated, the bibliography stops working. Here they are always left alone, and that behaviour is not a toggle: a DOI is never Uzbek.
Some journals accept articles in Latin script, some in Cyrillic, and some in either. The requirement differs from journal to journal, so check the rules of the publication you are targeting before you submit — the OAK-listed journal catalogues are linked below.
Make a habit of reading the converted text through once before you send it. The decisions list exists precisely for that: it does not ask you to re-read the whole text, only the handful of places where a call was made.
- Latin to Cyrillic and back
- Automatic script detection
- Every decision shown with its rule
- Links, DOIs and formulas left untouched
- DOCX conversion with formatting preserved
- The text stays on the device — conversion runs in the browser
Questions and answers
How does the converter work?
The text is converted inside your browser under the official 1995 alphabet rules. Input is repaired first (apostrophes, invisible characters, Unicode form), then the exception lexicon is checked, and finally the context-dependent rules run. Links, DOIs and formulas are never converted.
Do I have to register?
Not to convert: the result appears as you type, and you can read it, check it and copy it by hand with a mouse selection. A free account is needed only for the Copy and Download buttons. That is the only restriction on the page, and it does not depend on how long your text is.
What is the difference between the apostrophe and the tutuq mark?
They are two different characters. The mark in “oʻ” and “gʻ” is U+02BB and belongs to the letter itself. The tutuq belgisi in words like “sanʼat” or “eʼlon” is U+02BC. The plain keyboard ' is neither; the converter repairs it by position and reports that in the decisions list.
How are “Gʻ” and “Oʻ” written in Cyrillic?
Gʻ gʻ → Ғ ғ and Oʻ oʻ → Ў ў. Alongside them: Sh sh → Ш ш, Ch ch → Ч ч, Ng ng → Нг нг, Q q → Қ қ, H h → Ҳ ҳ, X x → Х х. “Ҳ” and “Х” are different letters, and confusing them is the most common Cyrillic typing mistake.
Can I use the result in a dissertation or an OAK-listed journal article?
Yes, but check the requirements of the specific journal before you submit: the script requirement differs between publications and no single general rule was found. It is worth reading the converted text through once — the decisions list is what makes that quick.
Domestic HAC Journals →How accurate is it, and how do I verify the result?
Around 90% of any text converts through direct mappings, where there is no room for error. The rest is decided by context. Press the marker above the result and the list shows every decision, the rule behind it, and the before and after forms.
Is the new 2019 / 2026 alphabet supported?
No. The converter implements only the 1995 alphabet that is currently in force. The new bill passed the Legislative Chamber on 7 July 2026 but has not cleared the Senate and has not been signed; on top of that, published sources disagree about which glyph is meant to replace “Oʻ”. Until the law is in force we will not add it on a guess.