iPDF.tools
EN

Language

OCR: Image and PDF to Text

5 languagesBeta

Extract text from images, screenshots and scanned PDFs in English, Chinese, Japanese and Korean. Copy it or save as TXT. Runs in your browser, no upload.

Files are not uploaded · processed in your browser

Image or PDF
or drop files here (files are processed locally, never uploaded)JPG, PNG, WebP and other images, or PDFs. Multiple files allowed (up to 50 pages, 50 files)
Languages

Up to 3. The first run downloads about 4.4 MB including the engine; after that it is cached.

Choose an image or PDF, select the languages and press "Extract text".

How to use

  1. Choose or drop images, screenshots or PDFs. You can select several at once.
  2. Select the languages that appear in the image, e.g. Japanese and English for mixed text.
  3. Press “Extract text”. The first time, the OCR data is downloaded and the progress bar shows each step.
  4. Copy the text or download it as a TXT file.

Tips for accurate results

  • Photograph text straight on, in focus, without glare or shadows.
  • Scan at around 300 dpi; grayscale is fine.
  • If you can already select the text in a PDF, PDF to Markdown Converter extracts it faster and more accurately.

Privacy

Files you choose and text you enter are processed in your browser. They are not uploaded to our servers and we do not store them. Some tools download the code or data they need (such as OCR language data) from this site the first time you use them.

Read the privacy policy

FAQ

Are my images or PDFs uploaded?

No. Recognition runs entirely in your browser and your files never leave your device. The first time, the page downloads the OCR engine and the language data you picked from this site.

How much does it download?

The engine is about 1.5 MB and each language is about 1.6–2.9 MB (English about 2.9 MB). Everything is cached by your browser, so later visits start right away.

Which languages are supported? Can it read mixed text?

English, Simplified Chinese, Traditional Chinese, Japanese and Korean. You can select up to 3 languages at once, for example Japanese + English for mixed documents. Each extra language makes recognition slower.

Which files can I use?

Common images (JPG, PNG, WebP, BMP, GIF) and PDFs, both scanned and digital. You can select several files, up to 50 pages in total. PDF pages are rendered at 300 dpi before recognition. PDFs that need a password to open must be unlocked first.

How can I get better results?

Use sharp, straight, evenly lit images; 300 dpi is ideal for scans. Only select the languages that actually appear. Handwriting, decorative fonts, vertical text and very small print are recognized poorly.

What does the recognized text look like?

Extra spaces between Chinese characters and Japanese kana, and extra blank lines, are removed; spaces between English or Korean words are kept. With several pages, each page starts with a page-number heading.

Can I use it on my phone?

Yes, it runs in your phone's browser. The first run downloads the recognition engine (about 1.5 MB) and each language you selected (about 1.6–2.9 MB), so use Wi-Fi if your mobile data is limited; after that your browser keeps them. Each file can be up to 200 MB, 50 pages in total, and recognition uses the device's processor, so it is slower on a phone.

Related tools