About
A lot of India still types in disguise.
A great many Hindi, Nepali, Bengali, Tamil and Kannada documents are really English letters wearing a special font. They look right on one computer and fall apart everywhere else. This site turns them into real text, and back again when a printer or a typing test insists.
Why it exists
Most converters are one find-and-replace.
Legacy Indian fonts store letters in the order they are drawn, while Unicode stores them in the order they are spoken. Swap characters one at a time and words like हिन्दी and कार्य come out broken: the vowel sign lands on the wrong side of the consonant.
We built the converter more like a compiler than a script: a tokenizer, a syllable reorderer and a test suite. Then we made it tell you what it couldn’t do, because a converter that silently damages a court order is worse than none.
It started with Hindi, Marathi, Nepali, Bengali, Tamil, Kannada, Gujarati, Punjabi, Malayalam, Telugu, Odia and 40 fonts, and grows one verified font at a time.
Principles
Four promises we can check.
- 01
Private by design
Conversion runs in your browser. We never see your text, which matters when it is a court record, an exam paper or a staff file.
- 02
Tested, not asserted
Every font ships with golden pairs, reorder tests and round-trip checks. If a font fails them, it doesn’t get a page.
- 03
Honest about gaps
Every font is compared with independent open-source converters, and every difference is published. When a font can’t show a character, you get a report instead of a silent substitution.
- 04
Free to use
The converters are free and need no sign-up. There are no ads inside or next to the tool.
How it’s built
A small engine, a lot of tests.
The converter is plain TypeScript with no dependencies, so the same code runs in the browser and in the test suite. This site is static pages generated from it.
- Tokenizer
- A longest-match tokenizer reads legacy text the way the font does, so two-key letters like “Hk” (भ) are never split.
- Syllable reorderer
- Visual-order marks such as the short-i sign and the reph are moved into Unicode order one syllable at a time.
- Profiles as data
- Each font is a mapping table plus a few rules. Pages, charts and this site’s tables are generated from those profiles.
- Diagnostics
- Anything unknown, lossy or malformed is reported with its position instead of being hidden.
Who runs it
Published as Indian Font Converters.
Indian Font Converters is an independent project that publishes this site at indianfontconverters.com. It is not affiliated with any font vendor, exam body or government department. Font names are trademarks of their respective owners.
We are responsible for every converter, page and claim here. If a conversion is wrong, a fact is out of date or a font is missing, tell us through the contact page and we will correct it. Corrections are tracked on the accuracy page.
The conversion engine is free software under the GPL-3.0-or-later. Its source code is available on request, and every mapping source and licence is listed on the licences page.
Where we are
Early, careful and open about it.
Each converter is compared with independent open-source converters, and the accuracy page lists every difference, including the ones still open. More fonts are in progress.