Translate scanned PDF free onlineBuilt-in OCR plus AI translation in 40+ languages
Upload a scanned or image-based PDF. The text is extracted in your browser using OCR, then translated. No server upload for OCR. No signup. No payment.
or drag and drop · maximum 50 MB · files are not stored
How scanned PDF translation works
From scanned image to translated text
Two steps, fully automated. Both run in your browser. Your scan is not uploaded for OCR.
1. Detect
The first pages of your PDF are checked. If no selectable text is found, the PDF is treated as scanned and OCR is offered.
2. Recognize
Tesseract OCR runs locally in your browser, page by page, in the language you select. Your file does not leave your device.
3. Translate
The extracted text is translated into your target language and shown next to the original scan, page by page.
Private OCR by default
Most online OCR-and-translate tools upload your scanned PDF to a server. Here, OCR uses WebAssembly Tesseract directly in your browser. The scan does not reach our servers. Only the recognized text is sent for translation.
Free, no quota
Most free online OCR services limit you to a few pages per day before requiring payment. Here, you can translate up to 15 pages per document, with no limit on the number of documents and no payment required.
40+ OCR languages
English, Arabic, Chinese, Japanese, Korean, Hindi, Marathi, Bengali, Tamil, Russian, Greek, Hebrew, Thai, and 30+ other languages. Select the language of your scan and the correct OCR model is loaded automatically.
No setup, no installation
Adobe Acrobat's OCR works well but costs $20 or more per month. Standalone Tesseract requires installation and command-line use. This page handles both with a single click.
FAQ
Translate scanned PDF — frequently asked questions
Can I translate a scanned PDF for free?+
Yes. OCR (text recognition) runs on your scanned PDF directly in your browser, then the extracted text is translated with AI. No signup, no watermark, no payment required.
How does scanned PDF translation work?+
Two steps. First, OCR extracts text from the page images. This happens in your browser using Tesseract, so the file does not leave your device. Then the extracted text is translated into your target language using a large language model.
Which languages does the OCR support?+
Tesseract supports more than 40 languages, including English, Spanish, French, German, Italian, Portuguese, Russian, Chinese (Simplified and Traditional), Japanese, Korean, Arabic, Hindi, Marathi, Bengali, Tamil, Turkish, Dutch, Polish, Czech, Greek, Hebrew, Thai, and Vietnamese.
How accurate is the OCR?+
Quality depends on the scan. Crisp 300 DPI scans of clean print typically come out nearly perfect. Faded text, handwriting, low-resolution photos of pages, or skewed scans can introduce recognition errors that propagate into the translation.
How long does it take?+
OCR runs locally in your browser, so it depends on your device. About 5 to 20 seconds per page for typical text. The translation step then takes a few seconds per page.
Are my scanned files uploaded anywhere?+
No. The PDF is rendered to a canvas and OCR is performed directly in your browser. The file itself is never sent to a server. Only the extracted text is sent for translation.
What is the page limit for scanned PDFs?+
The same as any other PDF: up to 15 pages and 50 MB free. OCR is more compute-intensive than text extraction, so longer documents will be slower. Splitting large files is recommended.
Will the translation keep the layout of the scan?+
Translation output is text-based. The visual scan is not rebuilt with translated text on top. You receive the translated text alongside the original PDF preview, which can be copied or exported.