extract text from scanned pdf

A regular PDF created from a word processor stores actual text characters on the page, which is exactly what a text-extraction tool reads. A scanned PDF is fundamentally different: it's a photograph or scanned image of a page saved inside a PDF file, with no text characters to extract at all, only pixels that happen to look like text.

This is why a standard text-extraction tool, including this one, returns empty or meaningless results on a scanned document. It's correctly reporting there's no text layer to read, not malfunctioning. Trying different extraction tools won't fix this, since the underlying problem is identical for all of them.

The correct technology for this job is OCR, optical character recognition, which analyzes the image itself and recognizes the shapes of letters and words even though there's no real text data behind them. This is a genuinely different process that needs a tool built specifically for it.

For a scanned PDF, the practical path is to run it through an OCR tool like Image to Text, which reads the text directly from the image and outputs it as editable, searchable text. Once you have that recognized text, you can use it the same way you would text extracted from a regular PDF.