Conservative. Idaho. Software engineer. Historian. Trying to prevent Idiocracy from becoming a documentary.
Email complaints/requests about copyright infringement to clayton @ claytoncramer.com. Reminder: the last copyright troll that bothered me went bankrupt.
"And we know that all things work together for good to them that love God, to them who are the called according to his purpose." -- Rom. 8:28
Pages
▼
Thursday, March 9, 2023
I Am Using ReadIRIS to OCR Historical Documents to Avoid Retyping Quotes (and Thereby Introduce Typos)
It works pretty well, but for this book, I am glad I bought a fast PC.
It seems to nne the "ChatAI" language models that scan words already used to predict the words intended tb be used next is well suited to copy edit OCR text. At the very least the "words' ' that seem to violate the AI's prediction could be flagged por human review. Say one line per page has a potential error flagged ... how much quicker to edit a entire book that to try reading and finding ALL fhe words that might have been misinterpreted by the Character Recognition softvvware?
Can you expand? Is it software that lets your pc's camera do ocr, or does it need separate hardware?
ReplyDeleteNo, it OCRs PDFs that are not searchable.
DeleteIt seems to nne the "ChatAI" language models that scan words already used to predict the words intended tb be used next is well suited to copy edit OCR text. At the very least the "words' ' that seem to violate the AI's prediction could be flagged por human review. Say one line per page has a potential error flagged ... how much quicker to edit a entire book that to try reading and finding ALL fhe words that might have been misinterpreted by the Character Recognition softvvware?
ReplyDelete