Transliteration & Language Tools
Convert writing from one script or notation to another by a named, published standard — Russian by ALA-LC or ISO 9, Arabic by ALA-LC or DIN 31635, Korean by Revised Romanization, English into the International Phonetic Alphabet — and see the rule behind each change. Free, no sign-up.
Language & Transliteration
23 tools · free, no sign-up
NATO Phonetic Alphabet Converter
Spell any text in the NATO phonetic alphabet (Alfa, Bravo, Charlie) — for spelling names over the phone, plus the full reference chart.
Braille Translator
Translate text to Grade 1 (uncontracted) Unicode braille and back — letters, numbers, and common punctuation, with the braille alphabet chart.
English to IPA Converter
Transcribe English text into the International Phonetic Alphabet with stress marks, word by word, from General American pronunciations in the CMU Pronouncing Dictionary.
Language Detector
Identify the language of any text among 59 — by its script where that settles it, by character trigrams where alphabets are shared — with confidence and runners-up.
Cyrillic to Latin Transliterator
Romanize Russian, Ukrainian, Belarusian and Bulgarian in ALA-LC, BGN/PCGN, ISO 9, the scholarly system and Russia's passport form, side by side with letter tables.
Greek to Latin Transliterator
Romanize Modern Greek in ELOT 743 and ALA-LC and Ancient Greek in the classical (SBL, Chicago) convention, with polytonic input, rough breathings and iota subscripts.
Korean Romanizer
Convert Hangul to the Revised Romanization of Korean with the official sound-change rules shown, or to its letter-by-letter transliteration, with name and hyphen options.
Pinyin Tone Converter
Turn numbered pinyin into tone marks (ni3 hao3 → nǐ hǎo, lv4 → lǜ) by the standard placement rule, convert tone marks back to numbers, or strip the tones for citations.
IPA Keyboard
Click-to-type International Phonetic Alphabet symbols laid out like the IPA chart, with consonants, vowels, diacritics, tones and tie bars, and a copy button.
Arabic Transliterator
Romanize Arabic script in ALA-LC, DMG/DIN 31635 and a simple form. Reads vowel marks, hyphenates the article, and says when unvowelled text hides the short vowels.
Hebrew Transliterator
Romanize Hebrew with or without niqqud in the SBL academic and general-purpose styles and Library of Congress ALA-LC, with spirant and aleph/ayin options.
Devanagari Transliterator
Convert Hindi and Sanskrit Devanagari to IAST or ISO 15919 Latin, and read both back into Devanagari, with anusvara, final-a and long e/o options.
Romaji Converter
Turn hiragana and katakana into revised Hepburn romaji with macrons, read romaji back into kana, and convert between hiragana and katakana.
Armenian Transliterator
Romanize Armenian in ISO 9985 and ALA-LC, Eastern or Western, with the prime and aspirate marks, and convert Latin text back to Armenian script.
Georgian Transliterator
Romanize Georgian in the 2002 national system, ISO 9984 and ALA-LC, and convert Latin text back to Georgian script, with a full letter table.
Serbian Cyrillic to Latin Converter
Convert Serbian Cyrillic to Gaj's Latin alphabet and back, handling lj, nj and dž both ways, plus Macedonian Cyrillic to Latin in two spellings.
Runes Translator
Spell Latin text in Elder Futhark, Younger Futhark or Anglo-Saxon Futhorc runes, and read runes back as letters. Transliteration, not translation.
Chinese Simplified to Traditional Converter
Convert Chinese between Simplified and Traditional characters with OpenCC tables, flagging every one-to-many character and offering Taiwan and Hong Kong forms.
Spanish Number Converter
Spells numbers as Spanish cardinal and ordinal words with gender agreement, apocope and the long scale, following the Real Academia Española.
French Number Converter
Spells numbers as French cardinal and ordinal words with the vingt and cent agreement rules, the 1990 hyphenation and the Belgian and Swiss variants.
German Number Converter
Spells numbers as German cardinal and ordinal words following the Duden, with compounding, capitalised large numbers, ein/eine/einen and Swiss spelling.
Chinese Number Converter
Converts Arabic numbers to Chinese numerals, financial 大写 numerals and RMB capital amounts, and reads Chinese numerals back into digits.
Hebrew Numeral Converter
Converts numbers from 1 to 9999 into Hebrew letter numerals with geresh, gershayim and the 15 and 16 convention, and reads Hebrew numerals back into numbers.
Romanization and Script Conversion
- Cyrillic and Greek. The Cyrillic to Latin Transliterator for Russian, Ukrainian, Belarusian and Bulgarian; the Serbian Cyrillic to Latin Converter, which works both ways in Gaj's Latin alphabet; and the Greek to Latin Transliterator for Modern and Ancient Greek.
- Semitic scripts. The Arabic Transliterator (ALA-LC, DMG/DIN 31635 or a simple form, with or without vowel marks) and the Hebrew Transliterator (SBL academic and general-purpose styles, or ALA-LC). The Hebrew Numeral Converter writes numbers as Hebrew letters.
- South Asia and the Caucasus. The Devanagari Transliterator converts Hindi and Sanskrit to and from IAST or ISO 15919; the Armenian and Georgian transliterators work in both directions.
- East Asia. The Korean Romanizer applies Revised Romanization, the Romaji Converter converts kana to Hepburn romaji and back, the Pinyin Tone Converter adds tone marks to Mandarin, and the Chinese Simplified to Traditional Converter flags every character with more than one traditional form.
- Old alphabets. The Runes Translator spells Latin text in Elder Futhark, Younger Futhark or Anglo-Saxon runes — transliteration, not translation.
Choosing a Romanization Standard
There is rarely one correct way to write a name in Latin letters. Libraries, governments, journals and passports each follow a standard, and the same Russian name can come out as Chaĭkovskiĭ, Chaykovskiy or Čajkovskij. Pick the system your context expects and use it consistently.
| If you are… | Usually follow |
|---|---|
| Citing a Russian, Ukrainian or Arabic source in a library catalog or a bibliography | ALA-LC — the Library of Congress tables |
| Writing a place name for English-language maps or government use | BGN/PCGN |
| Needing a reversible, one-to-one transliteration | ISO 9 for Cyrillic; ISO 15919 for Indic scripts |
| Writing Hebrew in biblical or academic work | The SBL academic style |
| Writing Sanskrit or Hindi in scholarship | IAST (or ISO 15919 for modern Indic languages) |
| Romanizing Korean for signs, maps or official documents | Revised Romanization of Korean |
| Writing Mandarin in Latin letters | Hanyu Pinyin with tone marks |
| Writing Japanese for English readers | Hepburn romaji with macrons |
When a person has an established spelling of their own name — on publications or a passport — use that spelling instead of a fresh transliteration.
Pronunciation, Spelling Aloud and Number Words
- Pronunciation. The English to IPA Converter transcribes English with stress marks and shows the dictionary entry behind each word, and the IPA Keyboard lets you type any symbol, laid out like the IPA chart.
- Other notations. The Braille Translator converts text to Grade 1 (uncontracted) braille and back, and the NATO Phonetic Alphabet Converter spells names and codes as Alfa, Bravo, Charlie for phone calls and radio.
- Numbers in other languages. The Spanish, French, German and Chinese number converters spell numbers in words by each language's own rules, such as the French 1990 hyphens and Belgian septante, or the German Duden rule that writes numbers below a million as one word.
- Identification. The Language Detector names the language of a passage from its script and letter patterns.
Accuracy and Limits
Transliteration converts letters, not meaning; none of these tools translates. Rule-based conversion is exact for the standard it names, but some standards depend on pronunciation or on vowels the spelling does not show — unvocalized Arabic and Hebrew, for example — so a romanized word can need a manual check against the rule the tool displays. The IPA converter uses the CMU Pronouncing Dictionary of North American English, so it reflects one accent and flags words it had to guess.
Most tools on the shelf run entirely in your browser. The IPA converter downloads its pronunciation dictionary once and looks up any word missing from it online, through the Datamuse API. For bibliographies in non-Latin scripts, see the guide to citing a foreign-language source.
Frequently Asked Questions
- What is the difference between transliteration and translation?
- Transliteration writes the same words in another script — Москва becomes Moskva. Translation changes the words into another language — Moscow. These tools transliterate; they do not translate.
- Which transliteration should I use in a bibliography?
- Follow your style guide or the library catalog record. In English-language scholarship, Russian, Arabic and other titles are usually romanized by the ALA-LC tables, which is what most library catalogs use. Whatever you choose, apply it to every entry.
- How do I transliterate Arabic or Hebrew without vowels?
- Arabic and Hebrew are normally written without short vowels, so a transliteration of unvocalized text has to leave them out or guess them. Add vowel marks (harakat or niqqud) first if you need an exact result; the transliterators handle text with and without them, and show which rule produced each letter.
- Does the IPA converter use British or American pronunciation?
- American. Its pronunciations come from the CMU Pronouncing Dictionary, which records General American English. British vowels, especially in words like bath and lot, will differ.
- How do I convert Simplified Chinese to Traditional Chinese?
- Paste the text into the Chinese Simplified to Traditional Converter. Character mapping is mostly one-to-one, but some simplified characters stand for several traditional ones (发 can be 發 or 髮), so the tool flags those for you to check, and offers Taiwan and Hong Kong forms.
- Is the braille translator suitable for producing materials for blind readers?
- It produces Grade 1 (uncontracted) Unicode braille, which is useful for learning, labels and checking. Most braille readers read contracted (Grade 2) braille, and printed materials should be produced and proofread by a certified transcriber.