thedashbetween

I have no issue with OCR parsing English, but when I try to scan a Greek document it never detects that it is in Greek. Instead it seems to detect Slovenian. I do have the Greek language loaded via the OCR languages parameter. If I override the default language to Greek then it parses the document in Greek without issue, but it is no longer able to parse English documents. Am I doing something wrong? Is this behavior expected? What if I have a document that contains both Greek and English. Some words, particularly proper nouns, are not always translated in Greek. So for those documents it tries to parse the English words as Greek (when Greek is forced as the default) and does so horribly. I understand OCR is not bulletproof, but it would be great to do as little manual work as possible when scanning in hundreds of documents. Thank you to whomever takes the time to read this and possibly offers help.