Forum Discussion
How to translate a PDF without copying text into a translator?
I’m reading some PDF documents in another language and need help understanding certain paragraphs as I go. Switching between the PDF and a translation website every time interrupts the reading, espec...
PyMuPDF + Argos Translate is an option if you want to translate selected passages locally. It requires some Python setup, but once the language model is installed, translation can run offline.
The workflow is:
- Install both libraries and download an Argos Translate model for your language pair.
- Use PyMuPDF to extract text from the PDF.
- Build a small reading window with the extracted text on one side and a translation panel on the other.
- Pass the selected passage to Argos Translate and display the result alongside it.
This avoids copying text into a website, but it’s a custom development approach, not a ready-made PDF reader. The extracted-text view won’t preserve the original page layout, and scanned pages need OCR first. It’s most useful if privacy matters and you’re comfortable setting up a small local tool.