PDF to Text

Pull the text out of reports, papers and e-books to quote, translate, summarise or search it; paste contract clauses into an email; feed documents to an AI assistant or a word counter; and archive the content of PDFs in a searchable text form.

A PDF stores text as positioned pieces on a page, not as paragraphs. The extractor puts the pieces back into lines from top to bottom and left to right, inserting spaces where there are gaps. In Paragraphs mode it then joins lines that belong together, starting a new paragraph where the gap between lines is clearly larger than usual, and removes the hyphen where a word was split across two lines.

Some PDFs contain pictures of pages rather than text — typically scans and photos of documents. They have no text to extract, so the result is empty or nearly empty. The tool detects this and tells you which pages have no text. For these, use PDF OCR, which recognises the text in the images. Mixed documents, where some pages are scanned, are common: the page-by-page message helps you find them.

Plain text cannot represent layout, so tables come out as lines with the cell values separated by spaces, and multi-column pages are read line by line across the columns where the columns are close together. Text in headers and footers is included with the rest of the page. Rotated or vertical text is left out. Fonts that embed text without a proper character map can produce unreadable characters — in that case OCR gives better results. To keep the layout, convert to Word with PDF to Word instead.

Contracts, bank statements and medical letters are private, so the PDF is processed on your device with pdf.js, the PDF engine of Firefox, and never uploaded. Files up to 300 MB can be opened, and long documents are read page by page with a progress indicator. Password-protected PDFs need to be unlocked first with Unlock PDF.