Aller au contenu
Retour à tous les outils№ 058

PDF en Markdown

Convertit le texte lisible en Markdown propre, prêt à coller dans des notes ou des docs.

0 B envoyés à un serveur

Ajoutez d’abord un fichier.

Copying text out of a PDF gives you a wall of words: the headings, the lists and the paragraph breaks are all gone, because none of them exist in the file as structure. This tool puts them back by measuring. It compares each line's font size against the page's body size — a line noticeably larger is a heading, and how much larger decides whether it becomes an H1, an H2 or an H3 — and it recognises bullet and number markers as lists.

That makes it a best-effort reconstruction, and the tool says so in the interface. A document with a consistent typographic hierarchy comes out remarkably clean; one that fakes its headings with bold body text will produce flat prose, because from the file's point of view that is what they are. Running text and paragraph breaks come through reliably either way.

How it works

  1. Drop in a PDF that contains real text.
  2. Read the extracted Markdown in the panel.
  3. Copy it to the clipboard, or download it as a .md file.

What people use it for

  • Getting a report into Obsidian, Notion or a wiki without retyping the structure.
  • Turning a PDF specification into a document you can diff and version in Git.
  • Pulling a long article into a notes app where you can actually annotate it.

Frequently asked questions

Is my file uploaded to a server?
No. The file is opened by JavaScript and WebAssembly running inside your own browser tab. It is never uploaded, there is no server-side processing and nothing is stored — closing the tab discards everything.
Nothing came out.
The PDF is a scan with no text layer. Run it through OCR first, then extract — there is no text to read until something writes one.
My headings came out as normal paragraphs.
They are probably not headings in the file — just body text set in bold, which is by far the most common way documents are written. The detector works on size, because size is what distinguishes a heading typographically. Add the # markers by hand where it matters.
Are tables converted to Markdown tables?
No. A PDF has no table structure to read, only aligned text, and guessing at cell boundaries here would produce tables that are subtly wrong. For tables use PDF to Excel, which does that inference deliberately and shows you the result in a grid.
What happens to images?
They are not included — this extracts text. Use Extract images to pull the pictures out separately and reference them from the Markdown yourself.

Outils similaires