Kruti Dev typing and conversion
Convert Hindi between Kruti Dev 010 and Unicode in both directions, and type Kruti Dev from an English keyboard. Runs entirely in your browser.
Unicode ↔ Kruti Dev converter
Kruti Dev is a legacy font: it stores Hindi as Latin byte codes, so the text only looks like Hindi when the Kruti Dev font is installed. Consonants, vowels, matras, the common conjuncts and digits are covered; rare conjuncts are not yet.
Or type Kruti Dev straight from an English keyboard
Press space after a word. Not the spelling you wanted? Pick another below or press 1–5.
Everything runs inside your browser. Your text is never uploaded, stored or sent anywhere.
Switch the output to Kruti Dev above to get legacy codes instead of Unicode.
What Kruti Dev actually stores
Kruti Dev is not an encoding of Hindi. It is a font that replaces the shapes of Latin
characters with Devanagari ones. When you type Hindi in Kruti Dev, the file records
ordinary Latin letters — नमस्ते is stored
as uelArs — and the font makes them look like Hindi on
screen. Remove the font and the Hindi disappears, because it was never there.
This is why Kruti Dev documents break when emailed, why they cannot be searched, and why pasting them into a web form produces gibberish. It is also why the conversion here has to be a real character-by-character mapping rather than a font change.
The reason anyone still uses it is inertia of a very practical kind: several state examination boards set typing papers in Kruti Dev, and offices hold years of material in it.
The matra that trips converters up
One rule accounts for most conversion errors you will see elsewhere. The vowel mark ि is drawn to the left of its consonant but, in Unicode, stored to the right of it. Kruti Dev has no reordering: it stores what you see, so the mark comes first. Convert without accounting for this and किताब comes back as कतिाब — close enough to look like a font problem rather than a conversion bug.
The converter here handles that reordering in both directions, and the round trip is checked automatically: text converted to Kruti Dev and back must come out identical to what went in.
What is not covered: consonants, vowels, matras, digits and the common conjuncts are mapped, but rare conjuncts are not. Anything unmapped is passed through unchanged rather than silently mangled, so you can see what did not convert.