Indic Script Converter

About the Indic Script Converter

Paste text in any of nine Indian scripts — Devanagari (Hindi, Marathi, Sanskrit, Nepali), Bengali–Assamese, Gurmukhi, Gujarati, Odia, Tamil, Telugu, Kannada or Malayalam — and get it back in another script or in Latin letters. The scripts descend from the same Brahmi writing system and Unicode encodes them in parallel, so most letters have an exact counterpart: संस्कृतम् becomes సంస్కృతమ్ in Telugu and ಸಂಸ್ಕೃತಮ್ in Kannada. Where a script lacks a letter, the converter writes the nearest one, follows that script’s own spelling rules and lists every such change, so nothing is altered silently.

For Latin letters you can choose IAST (the scholarly standard for Sanskrit), ISO 15919 (the international standard for all these scripts), ITRANS or Harvard-Kyoto (plain-keyboard schemes). All four also work the other way round: type dharmakṣetre or dharmakShetre and get धर्मक्षेत्रे. A separate Digits only mode turns Indian digits into 0–9, or 0–9 into any script’s digits, without touching the words.

Common uses

• Reading a Sanskrit verse, bhajan or name in the script you read best — Tamil, Telugu, Bengali or Latin.
• Writing Sanskrit and Hindi terms in IAST or ISO 15919 for papers, theses, library records and book indexes.
• Typing Devanagari on an ordinary keyboard with ITRANS or Harvard-Kyoto.
• Turning Indian digits in invoices, forms or scanned text into 0–9 for spreadsheets.
• Learning a second Indian script by comparing the same text in both.

How the conversion works

Each letter is read as one of a common set of sounds — the letters of ISO 15919, so क, ক, ਕ, க and k are all k — together with its vowel sign, virama (halant) or nasal sign, and written out again in the target. Letters are converted one by one, as transliteration standards require: the converter does not translate, and it keeps the spelling of the original rather than respelling words as they are pronounced in another language. Text in other scripts (for example English words in a Hindi sentence) is left as it is; text between ## markers in Latin input is not converted either.

Letters that do not exist in the target script

• Tamil has no separate letters for kh, g, gh, ch, jh, ṭh, ḍ, ḍh, th, d, dh, ph, b and bh. They are written with the nearest Tamil letter (भ → ப), or with superscript ² ³ ⁴ if you choose. Anusvara is written as the nasal it is pronounced as (शंकर → ஶங்கர), ऋ as ரு, visarga as ꞉, and न as ந at the start of a word or before த but ன elsewhere (भगवान् → பகவான்), following Tamil spelling.
• Gurmukhi has no ष or ऋ: ष becomes ਸ਼ and ऋ becomes ਰਿ. Anusvara becomes tippi ੰ after a, i and u and bindi ਂ after long vowels, and a doubled consonant takes the addak (सिक्ख → ਸਿੱਖ).
• Bengali–Assamese and Odia write v and b with one letter (व → ব, ବ; Odia ୱ after a consonant). य is য/ଯ at the start of a word and য়/ୟ after a vowel (समय → সময়). Going the other way, Bengali ব is read as b, but after a consonant as v (স্বাগত → स्वागत), except after ম and ব (লম্বা → लम्बा).
• Malayalam uses chillu letters for a final n, ṇ, r, l and ḷ and for r before a consonant (भगवान् → ഭഗവാൻ, धर्म → ധർമ).
• Bengali uses khanda ta ৎ for a final t (जगत् → জগৎ).

Every replacement of this kind is listed under Written differently with the reason, so you can check it.

IAST, ISO 15919, ITRANS and Harvard-Kyoto

IAST (from the International Congress of Orientalists, Geneva 1894) is the usual scheme for Sanskrit: ṛ for ऋ, ṃ for anusvara, e and o for ए and ओ. It has no letters for sounds outside Sanskrit, so this tool writes them as in ISO 15919, except ड़, ढ़ and ऴ, which are written r̤, r̤h and l̤ because IAST already uses ṛ for ऋ and ḻ for ळ.

ISO 15919:2001 covers all these scripts: r̥ for ऋ (ṛ is ड़), ṁ for anusvara (ṃ for the Gurmukhi tippi), ē and ō for ए and ओ so that plain e and o can be the short vowels of Tamil, Telugu, Kannada and Malayalam, and ṉ, ṟ, ḻ, ḵ for ன, ற, ழ, ஃ. It separates letters that would otherwise read as one with a colon (कइ → ka:i, पक्ह → pak:ha). Choose plain e and o for the standard’s non-uniform option for Hindi, Bengali, Gujarati, Punjabi and Odia. The options, notes and Gurmukhi rules used here follow A. P. Stone’s guide How to use ISO 15919 (archived copy).

ITRANS 5.3 and Harvard-Kyoto use only keyboard letters and are case-sensitive: A, I, U are the long vowels, T, D, N the retroflex consonants. ITRANS follows the encoding table at aczoom.com/itrans and accepts its alternatives (aa or A, sh, Sh, x for क्ष, GY for ज्ञ, .N for candrabindu); an underscore separates letters (क्ह → k_ha). Harvard-Kyoto writes ś as z and ṣ as S.

Limitations

• Transliteration is letter by letter. The optional Hindi-style rule drops only a word-final a; a silent a inside a word (कमला, kamlā) is still written.
• ITRANS and Harvard-Kyoto cannot write short e and o, ன ற ழ or (in Harvard-Kyoto) nukta letters; the nearest letter is used and listed. Use ISO 15919 for Tamil, Telugu, Kannada and Malayalam.
• Bengali and Odia do not distinguish b and v, and Tamil does not distinguish voiced and voiceless sounds, so converting to these scripts and back can change those letters.
• The rule that reads Bengali ব after a consonant as v fits Sanskrit words; a few words such as উদ্বোধন (ud-bodhan) are exceptions.
• Urdu (Perso-Arabic script), Sinhala, Grantha and Ol Chiki are not converted. Kruti Dev and other legacy-font text must first be turned into Unicode with the Kruti Dev to Unicode converter.
• Works on up to 2,000,000 characters at a time. If letters show as boxes, your device has no font for that script.

Privacy

Everything happens in your browser. What you enter or open here is not uploaded or stored by MySmartCoPilot.