Shusha ⇄ Unicode Converter

This page handles both Shusha conversion directions in a single tool. Paste Shusha-encoded Marathi or Hindi and get clean Unicode Devanagari — or start from Unicode text and produce Shusha ASCII for a legacy template. The Swap button between the two panes moves your text across, so you can run the opposite direction without retyping anything or hunting for a second page.

We keep both directions together because real Shusha work is rarely one-way. The typical job is a round trip: you pull text out of a 1990s-era Shusha document, clean it up in Unicode where spell-check and search actually work, and sometimes push the corrected version back into the original Shusha layout. One page means the mapping table, the troubleshooting notes, and the FAQ below apply to whichever way you are converting today.

Everything runs in your browser. No sign-up, no upload — the text never leaves your device, which matters when the document is a housing-society record, a court draft, or anything else you would rather not paste into an unknown server.

Shusha is a legacy ASCII Devanagari font developed by Harsh Kumar. Each Hindi or Marathi character is mapped to a Latin keystroke: type "k" with the Shusha font applied and the visual glyph is "क"; type "Aa" and the rendered output is "आ". Without Shusha installed, your device shows the underlying Latin characters, which is exactly why old Shusha files look like scrambled English on a modern machine.

• Step 1 — open the source document. Word files, PageMaker exports, and plain .txt files all work. Select only the Shusha-formatted passage (skip genuine English headings and names; see the mixed-font note under Troubleshooting) and copy it.
• Step 2 — paste into the left pane. The input box accepts up to ~5,000 characters per pass. The text will look like Latin gibberish in the box; that is normal; those are Shusha's raw ASCII codes.
• Step 3 — convert. The tool looks up each glyph in a mapping table sourced verbatim from the open-source hargup/unicode_to_shusha converter and emits the equivalent Unicode Devanagari sequence.
• Step 4 — proofread and copy. Skim the output for broken conjuncts, then click Copy output. The result pastes cleanly into Word, Gmail, government portals, WhatsApp, anywhere, with no font requirement on the destination machine.

The table covers ~57 entries: base consonants, the independent vowels, vowel matras, halant, anusvar, nukta. Rare stacked conjuncts may need a manual touch-up after conversion; the hargup source is canonical for the common Hindi/Marathi cases.

The reverse direction takes standard Unicode Devanagari — the text you type with Google Input Tools, an InScript keyboard, or any phone keyboard — and walks it back to Shusha's ASCII codes, character by character.

• Step 1 — bring in your Unicode Hindi or Marathi. Paste it fresh, or — if you are reversing an earlier conversion — click Swap to move the previous output across to the input pane.
• Step 2 — convert. The same ~57-entry table runs in reverse: each Devanagari code point is replaced with its Shusha keystroke sequence.
• Step 3 — expect Latin output. The result looks like random English letters. That is correct: Shusha text is only readable once the Shusha font is applied.
• Step 4 — apply the Shusha font. Paste the output into your document, select it, and set the font to Shusha. It will render as Devanagari. If it stays Latin, the font is not installed on that machine.

One thing no converter can do is install Shusha for the recipient. If the file is travelling onward, confirm the destination machine has the font — or send a PDF with fonts embedded so the rendering survives the trip.

Shusha → Unicode: rescuing old text

This is the direction most people arrive here for. Maharashtra produced an enormous amount of Shusha-typed material through the 1990s and 2000s: office orders, housing-society records, coaching-class notes, whole books laid out in PageMaker and CorelDRAW. All of it is locked to a font that modern systems do not ship. Converting to Unicode unlocks it:

• Old Marathi documents. A society resolution from 2003 or a coaching handout typed in Shusha opens as gibberish on any machine without the font. Convert it once and the text becomes permanent: searchable, editable, and readable on every device made since.
• DTP files moving to the web or Word. Text trapped in Shusha-era PageMaker and CorelDRAW layouts cannot be pasted into a website or a modern Word/Mangal workflow directly. Export the text, convert it here, and paste the Unicode result into your CMS or document.
• MPSC and Unicode exam prep. Maharashtra government typing has shifted to Unicode-first workflows, and older Shusha-era passages and question banks become reusable Mangal-compatible practice material once converted. For exam-realistic practice on the actual format, use our MPSC Marathi typing test.

Unicode → Shusha: feeding legacy workflows

Less common, but stubborn. Some offices and institutes standardised on Shusha twenty years ago and have never migrated, so anything you hand them has to arrive as Shusha bytes:

• Practice text for Shusha-based typing. Institutes that still train or test on the Shusha layout need fresh passages in Shusha form. Collect or type material in Unicode — far easier — then convert and hand over text that renders correctly under the Shusha font.
• Legacy templates. Letterheads, certificates, and DTP layouts built around Shusha only accept Shusha bytes. Converting your Unicode draft lets you keep the old artwork and replace just the text.
• Offices still on Shusha-era software. If the destination insists on Shusha, this direction lets a Unicode typist deliver without ever learning the legacy layout.

Shusha predates Unicode adoption in India. In the early 1990s there was no agreed way to store Devanagari on a computer, so font designers borrowed the only byte range every system understood: ASCII. A Shusha file does not contain Devanagari at all. It contains plain Latin bytes ("k", "Aa", "sa"), and the Shusha font simply paints Devanagari shapes over those bytes when it is applied.

The layout is broadly phonetic: k renders क, K renders ख, t renders त, sa renders स. That made Shusha far easier to pick up than Remington-style fonts such as Kruti Dev, where key positions follow the mechanics of a 1940s typewriter drum rather than sound, and it is a big part of why Shusha spread so widely through Marathi homes, schools, and government offices. (If your legacy text is Kruti Dev rather than Shusha, use our Kruti Dev → Unicode converter instead; the two encodings are not interchangeable.)

The same trick is also Shusha's fatal weakness. Because the file is "really" Latin text, nothing in it says Devanagari. Change the font and the meaning evaporates: paste Shusha text into Gmail, a web form, or any machine without the font, and you get raw keystrokes — gibberish. Search cannot find words, spell-check is useless, and screen readers announce English letters.

Unicode fixed this by giving every Devanagari character its own permanent code point (क is U+0915 wherever it goes). The character is stored as itself, so it renders identically on every modern device with no special font. That is why everything official (government portals, the Mangal font, modern exam software) runs on Unicode, and why for most documents conversion is a one-way migration: out of Shusha, into Unicode, for good.

Characters that don't map cleanly

The ~57-entry table covers the characters that appear in the overwhelming majority of Marathi and Hindi text. The fringe cases to watch:

• Stacked conjuncts. Jodakshar like क्ष, त्र, ज्ञ have variant representations in Shusha. Most convert correctly through the consonant + halant route; the occasional one comes out as separated components. Fix by retyping that cluster in Unicode.
• Less common signs. Characters outside the table — unusual Vedic marks, rare nukta consonants — pass through unchanged rather than converting wrongly. They are easy to spot in the output and patch by hand.
• Shusha variants. Shusha shipped as several font files over the years, and a few glyph slots differ between cuts. If your source used an uncommon variant, the odd character can map to the wrong glyph — always proofread before publishing.
• Numerals and punctuation. ASCII digits (0-9), Devanagari digits (० १ २ ३), and common punctuation pass through unchanged in both directions. If you need digit forms converted, do that manually.

Mixed-font documents

The most common source of mangled output is not the table — it is the input. Old Word files routinely alternate between Shusha passages and genuine English: names, addresses, dates, file numbers, headings. Because Shusha bytes are Latin bytes, no converter can tell the two apart. Feed it a real English sentence and it will faithfully "convert" it into Devanagari nonsense.

The fix is selection discipline. Convert the Shusha runs paragraph by paragraph and leave the English untouched, rather than pasting the whole document in one go. In Word, the Find pane with "Format → Font" can isolate every run formatted in Shusha, so you can work through them systematically instead of eyeballing.

Accuracy checklist

• Convert in chunks. Up to 5,000 characters per pass works best. Long pastes from PDFs often carry hidden formatting that confuses any converter; break the text into paragraphs.
• Proofread conjuncts. Skim the output for visibly broken stacks and fix them manually before the text goes anywhere official.
• Watch for diacritics. Anuswar (ं) converts reliably; if your source uses non-standard glyph mappings (older Shusha variants, alternate ASCII Devanagari fonts), signs like chandrabindu (ँ) and visarg (ः) may need touch-up.
• For exam answer scripts: never use a converter as a final layer. Practise typing in the actual format your exam ships: Shusha or Mangal for Maharashtra posts, Kruti Dev for SSC CHSL Hindi, Mangal/InScript for DSSSB and most state govt. The TypeForExam practice pages cover all of these.