PDF to Text Converter

Extracts text that is already stored in the PDF on this page.

Use this tool

Drop a file here, or use the picker. The file type is checked from its contents.

Result

The result appears in this panel after you run the tool.

About this tool

pdf-lib edits the page structure. Encrypted files are refused unless you are on the unlock page. Standard fonts are used when the tool draws text.

Scanned pages without a text layer stay empty. This tool does not perform OCR.

How to use it

  1. Open this page and provide the input it asks for.
  2. Set the options that are visible, then use Extract text.
  3. Read the result in the panel. Copy it or download it when a file is offered. The input stays in this browser.

Limitations

Reading the text layer the file already has

Stored strings, gathered per page

The extractor walks every page, from the first to the last. There is no page box and no scale, because nothing is rasterized. On each page it asks for the text content and keeps the string from each item that has one. Those strings are joined with a single space, and that line represents the page. Pages are then joined with a blank line between them, so the download is plain text with a gap where a page boundary was. Marked-content items that carry no string contribute an empty piece, and the spaces around that piece remain, which can leave a double space. The result is text, with a plain-text type, not a new PDF.

A scan without stored characters stays empty

Characters you can search in a viewer are stored as text operators and font references on the page. A photograph of a paper page is an image. This tool reads the stored text. It does not perform optical character recognition, so a scanned page with no text layer contributes an empty block for that page. The status line says the text was already stored in the PDF and that scanned pages are not recognized. If the whole file is a scan, expect little or nothing in the box. Rendering that page to JPEG or PNG will show you the picture. It will not invent a text layer either.

Order follows the items, not your eye

Items are emitted in the order the page stores them. A two-column article may come out with the columns interleaved, or with a running header sitting between sentences, because the extractor does not rebuild reading order. Hyphenation at line ends stays as the file stored it. Headers, footers, and ligatures come along with the rest of the strings. Copy the text into the text tools afterward if you need lines sorted, whitespace collapsed, or a word count. Encrypted files that the reader cannot open will fail before any page is listed. A file that mixes a scanned cover with a real text layer will show a thin gap for the cover and words for the pages that actually contain strings.

Privacy

Your input is processed in your browser and is not uploaded to UtilChain. File names, text, and results are not sent to analytics.