IAST (International Alphabet of Sanskrit Transliteration) is the academic standard for romanizing Devanagari and other Indic scripts. It uses Latin letters plus diacritics — macrons for long vowels, dots below for retroflex consonants — to represent every Sanskrit sound unambiguously.

How does the inherent vowel work?

In Devanagari every consonant carries an inherent short a. The tool adds that a after each consonant unless a different vowel sign or a virama (halant) follows. The virama removes the inherent vowel, giving a bare consonant cluster.

Why do some letters have dots or lines above them?

Those are the IAST diacritics: a macron (ā ī ū) marks a long vowel, a dot below (ṭ ḍ ṇ ṣ) marks a retroflex, ṛ marks vocalic r, ñ and ṅ mark palatal and velar nasals, and ḥ ṃ mark visarga and anusvara.

Is the conversion reversible?

Most well-formed Sanskrit and Hindi round-trips cleanly. The tool matches multi-character IAST sequences such as kh, ṣ and ai longest-first, and handles the inherent vowel and virama, so standard text converts both ways reliably.

No. The conversion uses a local lookup table in your browser, so nothing is sent to a server.

Devanagari ↔ IAST Transliterator

The Devanagari ↔ IAST transliterator converts between the Devanagari script (used for Sanskrit, Hindi, Marathi and Nepali) and IAST, the International Alphabet of Sanskrit Transliteration. IAST is the scholarly standard for romanizing Indic text, using diacritic marks so that each Sanskrit phoneme maps to exactly one Latin form.

How it works

Devanagari is an abugida: every consonant carries an inherent short a unless modified. The converter handles three cases. A consonant followed by a vowel sign (mātrā) takes that vowel — क + ि → ki. A consonant followed by a virama (्, the halant) loses its inherent vowel, producing a bare consonant for clusters — क् → k. A bare consonant with nothing after it gets the inherent vowel appended — क → ka. Independent (initial) vowels map directly: अ→a, आ→ā, ई→ī.

Going from IAST to Devanagari, the input is scanned longest-token-first so digraphs like kh, gh, ai, au and retroflexes are recognised before single letters. A consonant followed by another consonant gets a virama inserted between them to form the conjunct.

Example

The word “संस्कृत” (Sanskrit) transliterates to saṃskṛta. “योग” becomes yoga and “नमस्ते” becomes namaste. The retroflex in “कृष्ण” gives kṛṣṇa. Converting bhārata back returns “भारत”.

Notes

IAST distinguishes sounds English ignores: dental t (त) versus retroflex ṭ (ट), and short a versus long ā. Type the diacritics directly, or paste IAST from a source that already has them. Anusvara (ṃ), visarga (ḥ) and vocalic ṛ are all supported. Everything runs locally — your text is never uploaded.