Runs entirely in your browser - no upload, no sign-up.
Use this PDF page text extractor to read each page's selectable text into a separate, editable result. Move between pages, copy one page, or download every page with clear page markers. Processing stays in your browser, so the PDF is not uploaded. Image-only scans require OCR and are left empty.
Drop one PDF into the tool or select it from your device. PDFs with text you can select in a viewer work best.
Click Extract pages. PDF.js reads the text layer locally and reports progress as it finishes each source page.
Select a page in the navigator, compare its text with the PDF and correct any reading-order or character issues.
Copy the selected page, save it as its own TXT file, or export all pages with explicit page markers.
This tool uses Mozilla PDF.js to read the text layer page by page. It keeps blank pages in the navigator and in the combined export, so page numbers do not shift when a scan or separator page has no selectable text.
The result is plain text, not a visual reconstruction. Fonts, pictures, table borders and exact column geometry are not included. You can correct the extracted text in the page editor before copying it or saving UTF-8 TXT files.
No. The selected PDF is read locally by PDF.js inside your browser. CanDoYa does not receive the document or extracted text. The PDF engine code may load when you first start extraction, but your file is not included in that request.
Yes. You can extract, edit, copy and download page-separated text without an account, payment or watermark. Work runs on your device, so there is no server conversion queue and no uploaded document to retrieve or delete afterward.
The tool accepts one PDF up to 100 MB and 2,000 pages. Available memory and processor speed can create a lower practical limit, especially on a phone. Split unusually large documents into smaller PDFs if the browser cannot finish the extraction.
Not if the page contains only an image. This tool reads existing selectable text and does not perform optical character recognition. An image-only page remains in the navigator with a no-text label so later page numbers and combined export markers still match the source PDF.
A PDF often stores characters as positioned drawing instructions rather than paragraphs, columns or table cells. PDF.js follows the available text items, which may not match visual reading order on complex pages. Review each result against the source and edit it before reuse.
Yes. Select any page in the navigator, then use Copy page or Download page. The individual TXT filename includes the source page number. Copy all pages and Download all TXT create one combined result with an explicit marker before every page.
Yes, when the PDF includes a correct Unicode character map and selectable text. PDF.js can return text from many writing systems. Custom or missing font mappings may still produce incorrect characters, and image-only words require OCR regardless of language.
