Speaking as someone who has created PDFs of hundreds, perhaps thousands, of pages of documents and has edited the OCR text from many thousands of scanned pages of text (worked in Google's book scanning department for 2 years), the outlined process sounds about right if you require a low error rate./div>
Techdirt has not posted any stories submitted by Theo2.
(untitled comment)
Techdirt has not posted any stories submitted by Theo2.
Submit a story now.