Glyfo

Glyfo turns pictures of words back into words. Drag a box around part of the screen with Alt+Z, open an image or a PDF with Ctrl+O, paste from the clipboard, or send a file over from Explorer, Photos or Snipping Tool. A PDF can be read a page at a time or all at once; drop a folder of scans on the window and every one of them goes through in a single run. Save the result as .txt or .md, one file or one per page.

It also reads QR codes and barcodes out of the same picture, keeps a searchable history that survives a restart, reads the result aloud in any voice installed on your PC, and puts hard line breaks back into paragraphs — or strips the spaces that Chinese, Japanese and Korean text picks up during recognition.

Recognition uses the OCR engine that ships with Windows, on your machine. Glyfo makes no network connections at all: no account, no telemetry, no ads. Free and open source under the MIT licence.

It gives you the text, not a searchable PDF — no text layer is written back into the original file. Translation and the enhanced-accuracy recognition model require a Copilot+ PC.