Free or Low Cost OCR Solutions?
May 28, 2013 10:11 AM   Subscribe

Does anyone have any recommendations for a free or low cost solution that will let me scan to searchable PDF using OCR?

I just bought a new Epson XP 800 printer/scanner and while I like it, I am really disappointed that it did not come with OCR software. I run a paperless home office and would like to be able to scan directly to searchable PDF. I had gotten very used to doing this with my previous HP printer/scanner.
posted by soy_renfield to Computers & Internet (10 answers total) 5 users marked this as a favorite
posted by Lyn Never at 10:59 AM on May 28, 2013

OCR_Kadmos with irfanview?
posted by Obscure Reference at 11:06 AM on May 28, 2013 [1 favorite]

Mac or Windows?

Is there a card in the package, or an offer when you register the product online, for a discounted OCR package?

For waht it is worth, Microsoft Office, some versions, at least, comes with OCR functionality.
posted by Good Brain at 11:24 AM on May 28, 2013

As an FYI, you can use projects built from the Tesseract engine, which is free, but you get what you pay for. I've not been impressed by the quality. On top of that, you should be wary of the quality of output of some of the free PDF generation tools - I make a living writing tools that generate PDF (and I worked on Acrobat 1.0-4.0) and about 1/4 of my job is coping with shit PDF (I'm looking at *you* GhostScript).
posted by plinth at 11:37 AM on May 28, 2013

Nuance PDF isn't free, but it is amazing. I have been consistently impressed with OCR quality.
posted by Sheppagus at 12:29 PM on May 28, 2013

PDF-xchange viewer. Free, and OCRs like a dream. You would have to take the high-speed scanned image pdf and open it in this program to OCR. It's the only free PDF reader that OCRs that I have found.
posted by MeatheadBrokeMyChair at 1:22 PM on May 28, 2013

I scan to Evernote (premium edition) which automatically does OCR on the backend.

I *think* Google Drive also does this, although I'm not certain.
posted by ethidda at 2:07 PM on May 28, 2013

I used Abbyy Finereader Online to OCR some typewritten stuff, and I was very very happy with it the quality of the output—much better than Tesseract, OCRopus, and several commercial demos that I tried (better than Abbyy's desktop version, actually). It costs between $0.05 and $0.15 per page depending on how many pages you buy, so it may not be the best for you if you're planning an ongoing workflow.

I am also a heavy user of Evernote and I do use that for simple searchability: but for this particular project I needed to be able to edit the scanned text, and for that Abbyy had the cleanest output of the things I tried.
posted by xueexueg at 3:02 PM on May 28, 2013 [1 favorite]

This is on a Windows 8 machine. Thanks for all the suggestions! I'll try them out tomorrow.
posted by soy_renfield at 9:25 PM on May 28, 2013

So I finally got around to trying these suggestions. PDF-xchange viewer will OCR the documents, but it is a two step process. I have to scan the document and then open it in PDF Xchange, OCR it, and resave it. Given that it is free, I'll live with the extra steps.
posted by soy_renfield at 9:28 AM on June 11, 2013

« Older New baby + diss defense, can this math work?   |   What are these alchemical symbols on this pyramid? Newer »
This thread is closed to new comments.