I think I worded my inquiry poorly. Of course I do not want an OCR for a language dead for 2000 years! My problem I guess is that I have problems getting PDFelement to recognize UNICODE. As my example shows ša i-na bala-e Iiri-ba damar.utu ṭūḫḫeṣi; when PDFelement encounters a line like that in an article and creates a PDF, it seems to garble this line, even though it should be easily recognizable as UNICODE (Adobe PDF did), additionally it should be able to recognize superscripts and subscripts. I don't expect it to know what the line says (OCR), but it should be able to reproduce it in a PDF. Is there a tweak that takes PDFelement beyond ASCII with accents?