If you copy text out of an old Bangla PDF and get something like Avwg evsjv‡`‡ki instead of আমি বাংলাদেশের, the document was typed with a legacy Bangla font such as SutonnyMJ. The letters on screen are Latin characters drawn with a special font, so copying them gives gibberish. This tool reads the text out of the PDF and converts it to real Unicode Bangla that you can search, edit and paste anywhere.
It shows the raw text it read next to the converted text, tells you which fonts it found, and says so when a PDF has no text to read at all. It is a best-effort converter, not a guarantee: always read the result against the PDF.
সংক্ষেপে: বিজয় বা সুতন্বী এমজে ফন্টে তৈরি পিডিএফ থেকে লেখা বের করে ইউনিকোড বাংলায় রূপান্তর করে। আপনার ফাইল আপলোড হয় না; সব কাজ ব্রাউজারেই হয়।
Loading the tool…
Overview
About Bijoy / SutonnyMJ PDF to Unicode Text
Why legacy text looks like gibberish. Before Unicode was common, Bangla was typed with fonts such as Bijoy's SutonnyMJ that map Bangla letters onto the keys of an English keyboard. The file stores the English letters; only the font turns them into Bangla shapes. Anything that does not have that font (a search box, a database, another program) sees the English letters. Legacy text also stores letters in visual order: the vowel signs ি, ে and ৈ are written before the consonant, and the reph (র্) after the whole word, while Unicode stores them in logical order.
What this tool does. It reads the PDF with pdf.js and keeps each piece of text together with the name of its font. Pieces set in a known legacy font family (SutonnyMJ, other Bijoy "MJ" fonts and similar) are converted; ordinary English text in the same document is left as it is. If the font names are hidden, the tool looks for characters that only occur in legacy-encoded text and for very common legacy words, and converts the page when the evidence is strong. You can override this: convert everything that is not already Bangla Unicode, or turn conversion off and use the raw text.
The conversion. Text is read as tokens (single letters, and the combined glyphs legacy fonts use for conjuncts such as ক্ষ, ন্ত্র or স্ক্র), then each reph is moved in front of the consonant it belongs to, and each pre-consonant vowel sign is moved behind its consonant group, joining ে + া into ো and ে + ৗ into ৌ. The table was checked against several hundred words of ordinary Bangla news text, including conjuncts, reph, ya-phala and ra-phala. Digits can be converted to ০-৯, and "---- Page N ----" markers can be added.
Word boundaries. Spaces stored in the PDF are kept. Where a PDF stores words as separate pieces without a space, a space is added when the gap between pieces is wider than a normal letter gap. Lines and paragraph breaks follow the page layout.
Paste mode. If your PDF viewer lets you copy the legacy text, you can paste it into the box at the bottom and convert it without opening a PDF.
Compatibility
Supported Formats
PDF files up to 80 MB and 500 pages whose Bangla text is stored as real text in a Bijoy-style (ANSI) legacy font. Pasted legacy text is also accepted. PDFs with Bangla that is already Unicode are extracted as they are. PDFs that contain only images, or text converted to outlines, have no text to read.
Guide
How to Use Bijoy / SutonnyMJ PDF to Unicode Text
- Choose your PDF. Text is read in your browser.
- Check the report: it lists the fonts, how many pages have legacy or Unicode Bangla, and any page without text.
- Compare the raw text on the left with the Unicode text on the right. Edit the right-hand box if a word needs fixing.
- Copy the text or download it as a UTF-8 .txt file. Use "Convert all text" only if nothing was detected automatically.
Highlights
Features
- Raw and converted side by side See exactly what was read from the PDF and what it became, so mistakes are easy to spot.
- Font-aware Converts pieces set in legacy Bangla fonts and leaves English text in other fonts alone.
- Reph and vowel signs reordered Moves র্ in front of its consonant and ি ে ৈ behind theirs, and joins ো and ৌ.
- Stays on your device The PDF is read and converted in your browser; nothing is uploaded.
Applications
Use Cases
- Moving text from an old government circular, textbook or newspaper PDF into a Unicode document
- Making a legacy PDF searchable by copying its text into Word, Google Docs or a CMS
- Reusing a thesis or report typed in SutonnyMJ on a website
- Checking whether a Bangla PDF is legacy-encoded or already Unicode
- Converting a pasted block of Bijoy text without installing a keyboard or converter
Good to Know
Limitations
Not every legacy PDF converts perfectly. Legacy encodings were never standardised. Font variants, rarely used conjuncts, special symbols and files whose text was stored in an unusual order can come out wrong, and the converter prints characters it does not recognise unchanged and lists them. Always read the result against the PDF.
If the PDF is a scan, or the Bangla was converted to outlines (curves), there is no text in it. The tool tells you when it finds no text, and Bangla OCR is the next thing to try. Encrypted or password-protected PDFs must be unlocked first.
This tool runs entirely in your browser — your file is never uploaded to our servers. See our Privacy Policy for full details.
Support
Frequently Asked Questions
Is my PDF uploaded anywhere?
No. pdf.js reads the file in your browser tab and the conversion is a table lookup in the same tab. The text is not stored or sent to analytics.
Why does text copied from my PDF look like "Avwg evsjv"?
The PDF was typed with a legacy font such as SutonnyMJ, which stores English letters and draws them as Bangla. The letters Avwg are the stored form of আমি. Converting maps them back to Unicode.
Does it work with every Bijoy-style font?
It targets the common Bijoy ANSI layout used by SutonnyMJ and the other "MJ" fonts. Other legacy fonts can use different mappings, so check the result. If the font is not recognised by name, you can choose "Convert all text" and review the output.
The PDF is a scan or the letters are curves. What now?
Then no text exists to extract. The report says so. Use Bangla OCR, which reads text from a picture, and expect to proofread it.
Why are some words joined or split wrongly?
Some PDFs store each word, or even each letter, as a separate piece without a space between pieces. The tool adds a space when the gap is wide, but unusual layouts can still be mis-split. Edit the result in the box.
Can I use the result in Word or on a website?
Yes. Unicode Bangla works in any modern program and in web pages. You still need a font that has Bangla letters, such as Noto Sans Bengali, Nirmala UI or Kalpurush.
Keep Going
Related Tools
- PDF Information Read a PDF's page count, version, encryption flag and file size. The file is uploaded briefly and nothing is kept.
- PDF to Text Extract the real text layer from a PDF as a plain .txt file — free and fast.
- PDF Text Editor Click a line of text in your PDF, change or delete it, or add new text, then download the result. It all happens in your browser.
- Bangla OCR: Image to Bangla Text Read Bangla text from a photo or scan, in your browser. Copy or download the result as a text file, with a confidence estimate and tips for better results.
- Mushak-6.3 Tax Invoice / VAT Challan Generator মূসক-৬.৩ (কর চালানপত্র) ফরম্যাটে চালান তৈরি করুন: বিক্রেতা-ক্রেতার তথ্য, BIN, পণ্য, সম্পূরক শুল্ক, ভ্যাট, মোট ও কথায় টাকা। প্রিন্ট বা PDF। ব্রাউজারেই, নিবন্ধিত অনুমোদন ছাড়া।
- PDF Copy-Paste Text Health Checker Diagnose why text copied from a PDF comes out as symbols or wrong letters: ToUnicode maps, font encodings and a readable-text score for every page, with a copy preview. In your browser.