Nudi ↔ Unicode Converter

This page runs the full Nudi round trip. Paste glyph-encoded Nudi text and get Unicode Kannada that works in Word, Gmail, WhatsApp, and Karnataka government portals — or press the Swap button between the two boxes and feed it Unicode Kannada to get Nudi ASCII back for KGP-era templates and typing practice. The two directions used to live on separate pages; they now share this one tool, so a single bookmark covers either job.

Both directions use the same mapping table, roughly 250 entries ported from the canonical aravindavk/ascii2unicode converter — applied forwards for Nudi → Unicode and inverted, longest match first, for Unicode → Nudi. Conversion happens in your browser in well under 100 ms, and nothing is uploaded to any server.

Nudi is a glyph-encoded ASCII font family. Each Kannada letter or letter-piece is drawn on top of a Latin or extended-Latin code: type "PÀ" with a Nudi font applied and the screen shows ಕ; type "PÀ£ÀßqÀ" and you see ಕನ್ನಡ. The Kannada exists only in the font's artwork. Underneath, the file stores those Latin bytes, which is why a Nudi document opened on a phone, pasted into Gmail, or uploaded to a portal collapses into gibberish.

Unicode Kannada works the opposite way. Each character has its own code point — ಕ is U+0C95, ಡ is U+0CA1 — and renders identically on Windows, Android, iOS, macOS, and Linux with no font installation. Every modern website, portal, and messaging app expects it.

To convert in this direction:

• Step 1. Paste your Nudi text into the left box. The Paste button reads your clipboard directly.
• Step 2. Output appears on the right as you type; press Convert → to re-run it manually.
• Step 3. Skim the output for broken conjunct stacks, then press Copy output.
• Step 4. Paste anywhere — Word, Gmail, a government form. The receiving system needs no special font.

Behind the scenes the converter applies four mapping layers in priority order, ported from the open-source aravindavk/ascii2unicode Python converter: direct mappings (around 217 consonant-and-matra entries), arkavattu (the ð code that becomes the reph form of ರ), vattaksharagalu (subscript consonants that attach below a base letter), and broken cases (combining vowel signs that modify the previous character). Greedy longest-match-first scanning keeps multi-character sequences such as "ªÀiÁ" (ಮಾ) from being read as their shorter prefixes.

The reverse direction takes modern Unicode Kannada — text typed on a phone keyboard, Google Input Tools, KaGaPa phonetic, or Nudi 6.0 — and emits the ASCII codes that glyph-era Nudi workflows expect. Press the Swap button between the two boxes to flip the converter, paste your Unicode Kannada, and convert.

One thing surprises everyone the first time: the Nudi output looks like random Latin letters and accented symbols. That is correct behaviour. ಕನ್ನಡ becomes "PÀ£ÀßqÀ", and it stays that way until you paste it into a document and apply a Nudi glyph font, at which point it renders as ಕನ್ನಡ again. The converter produces the codes; the font produces the picture.

To convert in this direction:

• Step 1. Press Swap so the converter runs Unicode → Nudi.
• Step 2. Paste Unicode Kannada into the input box and convert.
• Step 3. Copy the ASCII output into Word, PageMaker, or your typing-practice software.
• Step 4. Select the pasted text and set the font to a Nudi glyph typeface (Nudi 01-e and its siblings). The Kannada appears.

The reverse table is built by inverting the same canonical mapping and sorting keys longest first, so compound sequences resolve before their prefixes. Vattaksharagalu and a few rare combinations have no single reverse code; proofread important documents in the exact Nudi font you will deliver in.

Nudi → Unicode: rescuing Karnataka's legacy files

Karnataka standardised on Nudi long before Unicode tooling matured, so two decades of official Kannada live in glyph encoding. Convert in this direction when that material has to work on modern systems:

• Government office files. Circulars, order sheets, and drafts typed in Nudi across secretariat and district offices turn to code the moment they leave the original PC. One conversion makes them readable, searchable, and shareable on any device.
• KGP-era documents headed for the web. Department notices and reports from the glyph-Nudi years need Unicode before they can go on a website, an e-office system, or a citizen portal. Search engines and screen readers cannot read glyph text at all.
• KPSC Unicode preparation. Coaching material for Karnataka exams circulated in Nudi for years. Convert old passages once and reuse them in our KPSC Kannada typing test, which runs in Unicode the way the actual computer-based test does.
• Baraha-era archives. Baraha shares most of its ASCII encoding with Nudi, so old Baraha exports usually convert with the same table.
• WhatsApp, Gmail, and the web generally. None of them render ASCII fonts. Unicode is the only Kannada that survives a share button.

Unicode → Nudi: feeding Nudi-only workflows

The reverse direction matters more than people expect, because parts of Karnataka's office and print ecosystem still run on glyph Nudi:

• Practice text for Nudi-based typing. Today's editorials and news copy are Unicode. Convert them to Nudi ASCII and load them into glyph-era practice software as fresh passages — useful if your office or institute still drills on the Nudi layout.
• Legacy office templates. Many taluk and district offices keep letterheads, proceedings formats, and registers set in Nudi glyph fonts. New text now arrives as Unicode from phones; converting it beats retyping it.
• Print shops and DTP. Wedding-card composers, book typesetters, and small presses across Karnataka built their page libraries in Nudi typefaces. Converting incoming Unicode keeps those libraries alive.

Nudi is the Karnataka government's Kannada standard. Developed by the Kannada Ganaka Parishat (KGP) and adopted by the state government in the early 2000s, Nudi became the default Kannada software in government offices, schools, and coaching institutes across Karnataka. That long head start is why so much official Kannada text exists in its encoding.

Glyph encoding, not characters. The classic Nudi typefaces (Nudi 01-e and the rest of the numbered family) store Kannada as ASCII art: the byte for "P" carries the picture of the bare ಕ shape, and accented codes like À, Á, and É act as vowel-sign modifiers. A single visible akshara is often two to four stored bytes — ಮಾ is "ªÀiÁ", three codes deep.

Vattaksharagalu and arkavattu have their own codes. Subscript consonants (the stacked second consonant in ಕನ್ನಡ or ಅಪ್ಪ) are separate glyph codes that attach to the previous base letter, and the arkavattu code ð becomes the ರ್ reph. This is why a converter needs layered logic rather than a flat character swap.

Versions changed the rules. Older KGP-era releases (through Nudi 4.0) are glyph-encoded and need this converter. Nudi 5.0 and especially Nudi 6.0 type Unicode natively, so text from them needs no conversion at all. If you are unsure which you have, paste a sample into the left box: glyph text looks like accented Latin, Unicode looks like Kannada.

Why Nudi text breaks outside Nudi fonts. A Unicode system reads "PÀ£ÀßqÀ" as seven Latin-range characters, because that is what the file contains. The Kannada only ever existed in the font's artwork. Spell-check, search, sorting, and screen readers all fail on glyph text; once converted, the characters themselves are Kannada and no font trickery is needed again.

Version mismatches

Files assembled over years often mix Nudi releases, or mix Nudi with Baraha export. The two share most of their encoding, but a handful of codepoints sit on different keys between variants. If a document converts cleanly for three paragraphs and then degrades, you have hit a font boundary — convert section by section and treat the odd section separately. Text that comes out half Kannada, half accented Latin almost always means the source mixed a glyph font with native Unicode; convert only the glyph portions.

Characters that do not map cleanly

• Rare conjunct stacks. Uncommon vattaksharagalu and triple-consonant clusters vary across Nudi versions; check ಕ್ಷ, ಜ್ಞ, ತ್ರ and similar clusters by eye after converting.
• The ಝ family. Nudi encodes ಝ with up to four ASCII codes per form ("gÀhÄ" and longer); broken sequences here are the most common manual fix.
• English mixed into Nudi text. The converter cannot tell English words from Nudi codes, since both are Latin letters. English inside a Nudi passage comes out as stray Kannada — strip it before converting and re-add it after.
• Numerals. ASCII digits (0-9) and Kannada digits (೦-೯) pass through unchanged. Convert digit styles by hand if your document needs one form.
• Formatting. Bold, italics, font size, and tables are lost; this is a plain-text converter. Reapply styling after pasting.

Accuracy checklist before you ship

• Convert in chunks of up to 5,000 characters; break long pastes into paragraphs.
• Copy from the original Word or text file when you can. PDFs reorder invisible characters and insert soft hyphens that confuse any converter.
• Spot-check anusvara (ಂ) and visarga (ಃ) near line ends, plus vowel signs on the last akshara of each line.
• For exam answer scripts, never rely on a converter at the final step. Practise typing in the layout your exam ships — our KPSC Kannada simulator covers both the Unicode and Nudi routes.