Getting the text
Choose a starting view
Plain text, Markdown, Outline or Tables. You can switch once the file is open.
Add the documents
Choose one or more PDF, Word, OpenDocument, RTF, text or Markdown files. PDFs show page progress while they are read.
Review the result
The figures show words, characters, pages or paragraphs and the detected format. Notes appear for scanned PDFs or misnamed files.
Take it away
Download TXT or MD, copy, or export tables or paragraphs to Excel or CSV. Several files can be saved together as a ZIP.
Tables and headings from a Word file
The same one-page Word test file in two views. Outline gives the heading structure, useful for checking that a long document is organised the way you think. Tables gives only the price table, separated by tabs, so it pastes straight into a spreadsheet as columns.
The Excel button would put that same table on its own sheet. For a document with several tables, each gets its own numbered section in the Tables view and its own sheet in the workbook.
terms-v1.docx: title, two subheadings, two paragraphs, one 2-column table
Outline: Wholesale terms 2027 Minimum order Payment Tables: Table 1 Range Trade price Stoneware mug £9.00 Pouring jug £16.00
Formats in this extractor
- DOCX, DOCM, DOTX, DOTM
- Word documents and templates from Word 2007 onwards. Headings, lists and tables are kept.
- DOC, DOT
- Word 97 to 2003. Text is read paragraph by paragraph; tab-separated runs of lines are treated as a table.
- ODT, OTT
- OpenDocument text from LibreOffice and similar suites.
- RTF
- Rich Text Format, produced by WordPad and many older systems.
- Read with pdf.js, page by page, with lines grouped into headings and paragraphs from the font sizes used.
When the text comes out wrong
PDFs store text as positioned fragments, not paragraphs, so the extractor rebuilds lines and paragraphs from where each piece sits. Two-column layouts, text boxes and footnotes can come out in an unexpected order. If that matters, the Plain text view is the best starting point for tidying by hand.
If you get almost nothing from a PDF, it is probably a scan. A note appears in that case; use an OCR tool first. For Word files, the text is read from the document body, so check the result against the document if it was heavily edited with tracked changes still in it.
Questions people ask
How do I extract text from a PDF?
Choose the PDF here and its text appears in the box, page by page, with headings and paragraphs rebuilt from the layout. Copy it or download it as a text or Markdown file. A note appears if the PDF looks scanned, in which case you need OCR to turn the page images into text first.
Can I keep the headings and tables?
Yes. The Markdown view writes headings with # marks and tables as Markdown tables, which keeps the structure in a form that pastes cleanly into notes, wikis and AI chats. The Tables view shows only the tables, tab separated, and the Excel export puts each table on a sheet.
Does it work with old .doc files?
Yes. Word 97 to 2003 documents are read as a series of paragraphs. Their formatting is stored very differently from modern files, so headings are not identified, but the text comes through, and lines that are separated by tabs are grouped as a table.
Will it extract text from Pages or Google Docs?
Not directly. Apple Pages files keep their text in a private format; export to DOCX or PDF from Pages first. Google Docs are not files on your computer until you download them; use File, Download and choose Word or PDF, then open that file here.
Where your data goes
Documents are read on your device. pdf.js and the other readers run inside this page, and no file or extracted text is sent to a server.
Checked against the tool on