PDF to PowerPoint (PPTX)
Best-effort pdf to powerpoint (pptx), from extracted text and a heading/list heuristic.
Runs on your device. The file is never uploaded.
PDF to PowerPoint extracts a document's text, guesses headings by glyph height, and lets pandoc's pptx writer build slides from that. A PDF holding no text stops with no extractable text found in this PDF. Slides break only where a heading was guessed, so a document set in one size gives the writer nothing to break on.
Questions
Does this send my file anywhere?
No. pdf.js pulls the text out in a Web Worker on your device, and pandoc compiled to WebAssembly writes the .pptx in the same worker. No upload, no sign-in, no copy kept. The tool is the same page you already downloaded, so it runs offline once you have visited once.
How does it decide where one slide ends and the next begins?
Indirectly, through the headings the extractor guessed. Text is turned into Markdown first: the most common glyph height counts as body text, and lines more than 15 percent taller become headings. Pandoc's pptx writer then builds the deck from that structure. A PDF set in one size throughout gives no headings, so there is nothing for the writer to break on.
My PDF was exported from PowerPoint. Will I get my deck back?
No. This is a one-way rebuild from extracted text, not a reversal of the original export. Slide layouts, master themes, images, speaker notes, transitions and animations are not in the PDF in any recoverable form, so they cannot come back. If you need the original deck, find the original file.
Why did it refuse with "no extractable text found in this PDF"?
Because there was no text layer to read. Scans, exported photos and image-only pages carry pixels, not characters, so the extractor comes back empty and the tool stops rather than writing you an empty deck. Run PDF OCR over the file first, then bring the OCR output back here.
What do I get back, and what is it named?
A single .pptx named after your PDF, so slides.pdf becomes slides.pptx. One file in, one file out, with no ZIP and no preview of the deck itself, since a pptx is binary. Your original PDF is left untouched on disk.
How do I keep the pages looking the way they do now?
Use PDF to JPG or PNG and place the images yourself. This tool takes the text path, so the visual appearance of each page is lost by design. Rendering each page to an image is the only route here that preserves layout, since it captures what the page draws rather than what it says.