Extract text from scanned PDFs and images using optical character recognition. Upload a scanned document or photo, select the language, and get editable text in seconds. Supports 14 languages. All processing happens in your browser.
or click to browse your files
Recognizing text...
Get editable text from any scanned document or image in three simple steps — no software installation required.
Drag and drop your scanned PDF or image into the upload area, or click "Choose File" to browse your device. The tool accepts PDF documents as well as PNG, JPEG, and WebP images. Whether your file is a multi-page scanned document or a single photograph of a page, it loads instantly in the browser so you can configure OCR settings right away.
Select the language of the text in your document for the best accuracy. Choose whether to process all pages or a specific range. Then click "Extract Text" to begin OCR recognition. The progress bar shows real-time status as each page is processed. Tesseract.js analyzes every pixel to recognize characters, words, and sentences.
Once OCR is complete, the recognized text appears in a scrollable text area with page separators. Review the content, then click "Copy to Clipboard" to paste it into any application, or click "Download as TXT" to save the text as a plain text file. Your original file is never modified or uploaded to any server.
A powerful, private, and free way to extract text from scanned documents — built for everyone.
There is no premium tier, no usage cap, and no hidden paywall. ConvertKr's OCR tool is free for everyone, whether you need to process a single scanned page or dozens of documents. We believe essential file tools should be accessible without a subscription or per-file charge.
Unlike most online OCR services, your files never leave your device. All text recognition happens locally in your browser using Tesseract.js. There is no server upload, no temporary cloud storage, and no risk of your confidential scanned documents being accessed by anyone else.
Recognize text in English, Spanish, French, German, Portuguese, Italian, Dutch, Arabic, Hindi, Chinese, Japanese, Korean, Russian, and Urdu. Select the correct language before processing to get the most accurate results from the Tesseract OCR engine.
A detailed progress bar and page counter show exactly how far along the OCR process is. You can see which page is currently being recognized and how many remain, so you always know when your text will be ready.
Upload scanned PDFs with multiple pages, or directly upload individual images in PNG, JPEG, or WebP format. The tool handles both workflows seamlessly — PDF pages are rendered to images internally before OCR processing begins.
ConvertKr runs entirely in your web browser, so it works on smartphones and tablets just as well as on desktop computers. There is nothing to install — if your device has a modern browser with JavaScript enabled, you can run OCR on the go. The interface is fully responsive and optimized for touch screens.
Everything you need to know about extracting text from scanned documents using OCR.
What Is OCR? OCR (Optical Character Recognition) is a technology that converts images of text into machine-readable text. When you scan a paper document, the result is an image — a photograph of the page. Even though you can see words and sentences in the image, a computer sees only pixels. OCR software analyzes the shapes of those pixels to identify individual letters, numbers, and symbols, then assembles them into words and sentences that you can edit, search, copy, and paste.
When Do You Need OCR? You need OCR when your PDF was created by scanning paper documents using a flatbed scanner, phone camera, or multifunction printer. These scanned PDFs contain images of pages rather than actual text data. If you try to select or copy text from a scanned PDF and nothing highlights, that means the document needs OCR. Common scenarios include scanned contracts, old archived documents, photographed receipts, book pages, handwritten notes, and any document that originated as physical paper.
How ConvertKr's OCR Works ConvertKr uses Tesseract.js, the JavaScript port of Google's open-source Tesseract OCR engine — one of the most accurate and widely used OCR engines in the world. When you upload a scanned PDF, the tool first renders each page as a high-resolution image using PDF.js. These images are then fed to the Tesseract OCR engine, which analyzes the visual patterns to recognize text. The engine uses trained language models to understand character shapes, word boundaries, and common language patterns. All of this runs entirely in your browser with no server involvement.
Tips for Best OCR Results The quality of OCR output depends heavily on the quality of the input image. For best results, use scans at 300 DPI or higher with good contrast between text and background. Avoid skewed or rotated pages — straighten them before scanning if possible. Select the correct language in the settings, as the OCR engine uses language-specific models to improve accuracy. Clean, clearly printed text in standard fonts will produce the best results. Handwritten text, decorative fonts, and very small text may reduce accuracy.
Privacy and Security All OCR processing happens entirely within your web browser. Your files are read into memory using the JavaScript FileReader API, rendered to images by PDF.js, and processed by Tesseract.js — all locally. The only network activity is downloading the Tesseract language data file from a CDN on first use, which is cached by your browser for subsequent sessions. At no point does your actual document data leave your device. This makes the tool safe for confidential documents, legal files, medical records, and any other sensitive content.
Everything you need to know about extracting text from scanned PDFs with ConvertKr.
OCR stands for Optical Character Recognition. It analyzes images of text — such as scanned documents or photographs — and converts the visual text into actual editable characters. ConvertKr uses Tesseract.js, the JavaScript port of Google's Tesseract engine, to perform recognition entirely in your browser without uploading files to any server.
ConvertKr supports 14 languages: English, Spanish, French, German, Portuguese, Italian, Dutch, Arabic, Hindi, Chinese (Simplified), Japanese, Korean, Russian, and Urdu. Select the correct language before processing for the most accurate results, as each language uses a specialized recognition model.
Accuracy depends on the quality of your source image. Clear, high-resolution scans (300 DPI or higher) with good contrast typically produce very accurate results. Blurry images, low resolution, unusual fonts, handwritten text, or complex layouts with tables and columns may reduce accuracy. Clean printed text in standard fonts works best.
Yes. After uploading your PDF, you can choose to process all pages or specify a custom page range. This is especially useful for large scanned documents where you only need text from certain pages, since OCR processing takes several seconds per page depending on your device.
Absolutely. All OCR processing happens entirely in your browser using Tesseract.js. Your files are never uploaded to any server, and we cannot see, access, or store your documents. The only network request is downloading the language data on first use. Once you close the tab, all data is cleared from memory.
Regular text extraction reads embedded text data from a PDF, which is nearly instant. OCR must analyze every pixel of an image to recognize characters — a much more intensive process. Each page typically takes a few seconds depending on your device. The progress bar shows real-time status so you always know how far along the process is.
Explore other free tools from ConvertKr to handle all your file needs.
Extract embedded text from PDF documents (non-scanned).
Open Tool →Combine multiple PDF files into a single document with drag-to-reorder.
Open Tool →Reduce PDF file size by re-rendering pages as optimized images.
Open Tool →Extract page ranges from a PDF into separate files.
Open Tool →Convert PDF pages to high-quality PNG or JPEG images.
Open Tool →Add text, images, signatures, drawings, and redactions to any PDF file.
Open Tool →