Could it just be that bootleg operations are scanning in the books with cameras rigs, and then using tesseract or some similar OCR to parse the text back out an

Could it just be that bootleg operations are scanning in the books with cameras rigs, and then using tesseract or some similar OCR to parse the text back out and then running it through Microsoft Word spell check to manually resolve conversion errors?