Hindi to English transliteration in IAST, ISO 15919, ITRANS and Harvard-Kyoto
This Devanagari transliterator writes Hindi, Sanskrit, Marathi or Nepali text in Roman (English) letters: कृष्ण becomes kṛṣṇa in IAST, kr̥ṣṇa in ISO 15919, kRRiShNa in ITRANS and kRSNa in Harvard-Kyoto. It also works the other way, turning any of those four schemes back into Devanagari, so it doubles as an IAST or ITRANS to Hindi converter.
Students quoting a verse, researchers preparing a bibliography, editors who need a name in a library catalogue and anyone who wants to read Hindi text without knowing the script can use it. A Hindi style drops the silent a the way Hindi is spoken (भारत → bhārat), while Sanskrit style keeps every vowel and converts back letter for letter.
The conversion runs entirely in your browser, and a badge under the result tells you whether the transliteration converts back to exactly the text you started with.
How to use the Devanagari Transliterator
- Paste or type Devanagari in the left box (the page opens with a Hindi example).
- Choose a Scheme: IAST for academic writing, ISO 15919 for library and catalogue work, ITRANS or Harvard-Kyoto for plain keyboard-only text.
- Pick Hindi for natural modern Hindi or Sanskrit for an exact, reversible transliteration.
- Check the badge and the Word by word table, which shows each word in all four schemes.
- Copy or download the result, or press Use as input to convert it back the other way.
Devanagari Transliterator features
Four standard schemes
IAST (Geneva, 1894), ISO 15919 (2001), ITRANS 5.2 and Harvard-Kyoto, with a full letter table in the tool.
Both directions
Roman → Devanagari accepts every listed spelling of each letter, including ITRANS alternatives such as aa/A and ISO text with or without macrons on e and o.
Exact round trips
In Sanskrit style every conjunct, nukta letter, reph, avagraha, visarga and zero-width joiner survives the trip there and back, checked live.
Hindi pronunciation option
Heuristic schwa deletion plus class nasals (हिंदी → hindī) and tilde nasal vowels (माँ → mā̃) as ISO 15919 recommends for modern languages.
Unambiguous output
Where two letters would read as one, a separator is added: कइ → ka:i, क्ह → k:ha.
Which scheme should you use?
Sources, checked 19 October 2026: IAST as adopted by the Transliteration Committee of the Tenth International Congress of Orientalists (Geneva, September 1894); ISO 15919:2001 tables; the ITRANS 5.2 encoding table published by sanskritdocuments.org (last updated 31 March 2000; current ITRANS release 5.34, April 2013); Harvard-Kyoto as tabulated by Wikipedia from Wujastyk (1996).
| Scheme | Letters | Best for | Example (कृष्ण शास्त्री) |
|---|---|---|---|
| IAST | diacritics (ā ṣ ṇ ś ṛ) | Sanskrit and Indology publications | kṛṣṇa śāstrī |
| ISO 15919 | diacritics, ē ō, r̥ | Libraries, catalogues, all Indic scripts | kr̥ṣṇa śāstrī |
| ITRANS | plain ASCII, case-sensitive | Typing on any keyboard, e-texts | kRRiShNa shAstrI |
| Harvard-Kyoto | plain ASCII, one letter per sound | Compact search-friendly text | kRSNa zAstrI |
How the Hindi silent-a option works
Every Devanagari consonant carries an inherent a, but Hindi does not pronounce it at the end of most words or in the middle of many: कमला is said kamlā, not kamalā. The Hindi style applies the standard right-to-left rule described in phonology studies of Hindi: a final inherent a is dropped (but kept after a conjunct ending in y, r, l or v, as in मित्र mitra), and a medial one is dropped when a vowel-and-consonant comes before it and a consonant-and-vowel after it.
The rule cannot see word boundaries inside compounds or recognise loanwords, so treat Hindi output as a reading aid. For citations and anything you may convert back, use Sanskrit style.
Letters the schemes do not cover
- IAST and Harvard-Kyoto were designed for Sanskrit and have no letters for the nukta sounds of Hindi and Urdu loanwords (क़ ख़ ग़ ज़ फ़ ड़ ढ़). This tool uses ISO 15919 letters in IAST (q, k͟h, ġ, z, f), writes ड़/ढ़ as ṙ/ṙh because IAST ṛ is already ऋ, and in Harvard-Kyoto puts a dot before the letter (
.D, as ITRANS does). - ॐ is written oṃ in IAST; a lone oṃ converts back to ॐ. Devanagari digits become 0–9.
- To type new Hindi from English letters use Hindi typing; to transcribe English pronunciation use the IPA converter; to inspect the code points use text to Unicode.
Devanagari Transliterator FAQ
How do I convert Hindi to English letters?
Paste the Hindi text, choose a scheme (IAST is the most widely read) and Hindi style, then copy the result. That is transliteration: the words stay Hindi, only the letters change.
Is transliteration the same as translation?
No. Transliteration in English letters keeps the sound (नमस्ते → namaste); translation changes the meaning into English (hello). Use a translator for meaning.
How do I convert Sanskrit to Hindi?
Sanskrit and Hindi share the Devanagari script, so Sanskrit text needs no script conversion. If you have Sanskrit in Roman letters (IAST, ITRANS or Harvard-Kyoto), choose Roman → Devanagari and the matching scheme. Translating Sanskrit meaning into Hindi is a different task.
What is the difference between IAST and ISO 15919?
They agree on most letters. ISO writes ṁ for anusvara, ē and ō for the long e and o, and r̥ for vocalic r, which frees ṛ for the Hindi letter ड़. IAST writes ṃ, e, o and ṛ.
Why does my text show a colon or underscore?
It marks a break where two letters would otherwise be read as one: ka:i is क + इ, while kai is कै. Keep it if you want to convert back exactly.
Is my text uploaded?
No. The tables and rules run in your browser.