Extract text from a PDF

How it works

Choose a PDF and you will see the text of each page separately, with a button to copy each one or all of them. With the create button you download a .txt file in UTF-8, with or without a line marking the start of each page. You can limit it to some pages (for example 1-3, 5).

Only the text the PDF carries as text is extracted. If the PDF is a scan or a photo of a document, each page is an image and there is no text to pull out: in that case the tool tells you so and sends you to “OCR PDF”, which recognizes the text in images.

You can choose whether the text of form fields and annotations is included, and join words split by a hyphen at the end of a line. The position of each word is not shown: this is plain text only.

What is not preserved, and limits

• The reading order is an estimate: the text in a PDF does not say in what order it is read, so it is arranged by lines and blocks, top to bottom and left to right. In PDFs with several columns, tables, footers or text in boxes it may come out in a different order or mixed.
• A scanned PDF, with no text layer, gives no text. Some poorly defined fonts give wrong or missing characters when copying, as also happens in a PDF reader.
• Formatting is lost: bold, sizes, tables, images and the position of the text are not kept. Tables come out as lines of text, not as cells.
• Text in other scripts (Arabic, Hebrew, Indic scripts…) is extracted as the PDF stores it, which may not be in logical order; no right-to-left reordering is done.
• Password-protected PDFs can be processed: the site asks for the password and it is only used in your browser, never saved or sent. The result is saved without a password. PDFs protected with a digital certificate instead of a password cannot be opened.
• Very large files (over roughly 100-150 MB in total, or with thousands of pages) may fail, especially on phones, because everything happens in your device's memory.