1. 1Add pages
  2. 2Read
  3. 3Style & download

1Add pages

    Each photo you take is added as the next page.

    2Read

    Your photos are read on this device and never uploaded. The reading engine and language data (about 5 to 10 MB) download once from jsDelivr and are then cached by your browser.

    Step by step with the Handwriting to Text Converter

    1. Add photos of your handwritten pages: choose them, drag them in, paste them, or use Take a photo on a phone. Up to 50 pages.
    2. Put the pages in order by dragging the list or using the arrows.
    3. Choose the language the pages are written in, and a second one if they mix two, then select Read handwriting.
    4. Check each page's text and correct anything the reader got wrong. To get the document in another language, choose it under Document language and select Translate.
    5. Choose the format (PDF, Word DOCX or DOC, ODT or TXT), font, size, text colour, page colour, spacing, alignment and page size, then download.

    A real example

    A two-page set of class notes in Hindi and English, photographed with a phone:

    1. Add both photos, or take them with the camera button, and check the pages are in the right order.
    2. Choose Hindi under Handwriting is in and English under Also contains, then select Read handwriting.
    3. Correct any words the reader misread in each page's text box.
    4. Choose Word (DOCX), the Nirmala UI font at 14 pt, a cream page colour and 1.5 line spacing, and download.

    The result is an editable Word document with two pages, the Hindi text marked as Hindi so Word's spelling tools and screen readers treat it correctly.

    What happens to your image

    The photos are read by Tesseract, an open-source text recognition engine, running inside your browser. Nothing you add is uploaded: each photo is decoded, straightened by its camera orientation, downscaled to a size the engine reads well and, with clean-up switched on, turned grey and its contrast stretched so pale ink stands out from the paper.

    Several pages are read at the same time, one per processor core of your device, so a batch of photos finishes much sooner than reading them one after another. The engine and the language data download once from jsDelivr, a free public CDN, and your browser keeps them, so later visits start straight away.

    Tesseract reads print and neat block handwriting well. Joined-up cursive is harder for it, which is why every page's text is shown for you to check and correct before anything is saved. The document is then built on your device in the format you choose, with your font, size, colours, spacing, alignment, page size and margins.

    Word (DOCX), OpenDocument (ODT) and plain text keep the text editable. PDF pages are drawn in your chosen font exactly as you see them, which keeps Hindi, Arabic and Chinese text correct in any PDF reader. The DOC option is the format Word saves as a web page, which every version of Word opens as a document.

    To get the document in another language, choose it under Document language and select Translate. Recent Chrome and Edge translate on your device with their built-in translator. Other browsers use a free open-source translation model that runs in the page and downloads once from Hugging Face: a small Opus-MT model (about 100 MB) where one exists for the pair, such as English and Hindi, and otherwise Meta’s NLLB-200 (about 850 MB), which covers every language listed, including Gujarati, Tamil and Bengali. Either way your text never leaves your device.

    When the Handwriting to Text Converter helps

    • Turning class notes or lecture notes into a Word document you can edit, search and share.
    • Typing up a handwritten letter, diary or recipe without retyping it by hand.
    • Converting handwritten Hindi, Marathi, Bengali or other Indian-language pages into editable text.
    • Making a clean, printable PDF of a handwritten form, list or draft.
    • Digitising a notebook that must not be uploaded to an online service.

    If the result is not what you expected

    The text is mostly wrong or empty
    Photograph the page straight on, in even daylight, filling the frame, and keep Clean up photos switched on. Tesseract reads print and neat block handwriting well but struggles with joined-up cursive and very small writing; for those, correct the text in the page's box before downloading.
    Reading never starts or stops at loading the engine
    The reading engine and language data download from jsDelivr on first use, about 5 to 10 MB. Check the connection, and that a network filter or ad blocker is not blocking cdn.jsdelivr.net. After the first successful run your browser keeps them, so it starts quickly afterwards.
    Hindi or other Indian-language text shows as boxes in Word
    The font chosen for the document does not contain that script on the computer opening it. Choose Nirmala UI (Windows), Mangal or Noto Sans Devanagari before downloading, or change the font in Word. The PDF format draws the text itself, so it always looks the same.
    The pages come out in the wrong order
    The document follows the order of the photo list, not the order the files were taken or named. Drag the photos into the right order, or use each photo's arrows, then select Read handwriting again.

    Questions about the Handwriting to Text Converter

    Is this handwriting to text converter free?

    Yes. It is free with no sign-up, no watermark and no page limit per day, because everything runs in your browser. You can convert up to 50 pages at a time.

    Are my photos uploaded?

    No. Your photos are read on your own device and the document is created there too. The only downloads are the reading engine and language data, fetched once from jsDelivr, a public CDN; your photos and text never leave the page.

    Can it read cursive handwriting?

    It reads print and neat block handwriting well, and joined-up cursive less reliably. Every page's text is shown for you to check and correct before you download, so the document ends up exactly as you want it.

    Which languages can it read?

    English, Hindi, Marathi, Bengali, Gujarati, Punjabi, Tamil, Telugu, Kannada, Malayalam, Odia, Nepali, Urdu, Arabic, Spanish, French, German, Portuguese, Italian, Russian, Simplified Chinese and Japanese. You can choose two languages for pages that mix them, such as Hindi and English.

    What is the difference between DOCX and DOC?

    DOCX is the modern Word format and the best choice for editing. DOC is offered for older Word versions: it is the format Word saves as a web page, which every version of Word opens as a document. Both keep your font, colours and page settings.

    Can I use it on my phone?

    Yes. On a phone, Take a photo opens the camera, and each photo you take is added as the next page. Reading runs on the phone itself, so newer phones are faster.

    Can it translate my notes into another language?

    Yes. After reading, choose a language under Document language and select Translate: any of the languages it reads can be translated into any other, such as English notes into Gujarati or Hindi. Recent Chrome and Edge use their built-in translator. Other browsers download a free translation model once, about 100 MB for common pairs such as English and Hindi and about 850 MB for Meta's NLLB model, which covers the rest. The translation runs on your device, and Show original text switches back.