When you use Google Translate to translate a PDF document made searchable by an OCR tool from a PDF image, the translated text that is generated is well positioned but overwritten on the original images and, as such, unreadable. Many discussions on the Google Translate forum were opened and are now closed without any solution (like using Adobe PDF Reader which does not work etc).
A first solution consists in using
$ pdftotext text_output_from_google_translate.pdf
and the result is a text file text_output_from_google_translate.txt .
pdftotext comes from the poppler suite but, and here is my question:
"How can I approximately preserve the positions and pages of translated text that was positioned on the pdf of Google Translate but overlaid?"