Thank you for your review
$68 20
Email me when discounted: 

OCRKit uses OCR (Optical Character Recognition) technology to recognize the text in the graphic, which is particularly useful for PDFs received via e-mail, created by DTP, office applications, or with a scanner or copier. Create searchable PDFs, RTF, HTML, or Unicode plain text simply by dragging a document on the OCRKit application or Dock icon.

Searchable Text. OCRKit is a simple and streamlined Mac application, that features the advanced Optical Character Recognition technology, allowing you to convert scanned or printed documents into searchable and editable text. It can be a great help for more...

What's New

Version 2.6: Release notes were unavailable when this listing was updated.

Note: Version 2.6 is available only to MacUpdate Desktop users. Version 2.5 is still the latest demo available to the public.


OS X 10.5 or later

Similar Software

Open Comparison
Suggest Other Similar Software
Leave a Review

OCRKit User Discussion

Nobody has reviewed or commented on this app yet. Add your own comment and get a discussion going!
Sort by: Time | Smiles
Aeschylus Member IconReview+158

OCR Kit does as good a job as any OCR product on the Mac. Compared to using Adobe Acrobat on a PC, its accuracy is similar, although its features are less. Comparing prices, there's no comparison. While it may take a few minutes to learn OCR kit, there is little to do for single document conversion and it is very reliable. I like it a lot, and it is the only reason I can do OCR on my Mac. Thanks for a nice product.

Reply1 reply
Version 2.4.1

Can you or someone tell how good is it with german. Can it learn new words as it OCRs? Have a user dict to edit? Thanks

Ryde Member IconReview+105

I've purchased quite a few licenses for OCR applications in the last couple of months, and all of these products have something going for them, but OCRKit stands out for at least two reasons:

a) OCRKit can be invoked in an automated, i.e. "batch", fashion.
b) OCRKit is able to preserve CCITT Group 4 compression in scanned images, at least in some cases. This finally gives me output files which have the exact same image quality as the original and are only marginally bigger due to the added text content. Kudos!

OCRKit is not perfectly straightforward to use IMHO and could definitely use a better documentation, though.

Regarding the recognition accuracy: I have been able to obtain very good results with OCRKit provided that the quality of the scanned image is high. In other cases the recognized text will contain errors, as with any other OCR software. OCRKit is said to be able to utilize a dictionary for better results, but it seems to me that this does not yet work in all cases.

So overall I am quite impressed with OCRKit, but improvements are of cause welcome. I hope the developers continue to bring their app forward, it has a lot of potential.

Please also see my comments added to the review of "brianvon" right below.

Reply0 replies
Version 1.13
brianvon Member IconReview-2

Bloatware, takes my 1 MB file and at MINIMUM file size expands it to nearly 10 MB. Under higher quality settings it bloats the pdf 33x larger.

Spell correction does not work well. It took an easily readable word "confirmation" --> "confirmatioIL"

And that was with dictionary turned on!

try again

Reply1 reply
Version 1.13

Unfortunately, your findings do apply to most OCR available on the Mac, not just OCRKit (which does actually better than many competitors, see below). So your half-star rating is a too bit harsh IMO.

The problem is that most Mac applications which deal with PDFs are implemented using Apples PDFKit framework which has some limitations. In particular, when writing out PDFs, it can only write a subset of the image codecs defined by the PDF standard. So if you got PDFs which use a sophisticated special purpose (i.e. high-quality and high-compression) image codec like JBIG2 or CCITT Group 4, your average PDFKit based OCR application which only supports Deflate and JPEG as its compression methods has a problem. Either will it create files which are much bigger than the original (Deflate keeps the image quality as is, but is too simplistic to compress well) or which got degraded graphics (JPEG trades image quality for smaller file sizes). To avoid these drawbacks, the software developers would either need to go extra miles and partly bypass PDFKit or use a different PDF handling library altogether. Both requires effort, knowledge and maybe additional licensing costs, so most companies don't seem to bother. The developers of OCRKit are a notable exception, because at least in some cases OCRKit is actually able to transfer the images 1:1.

Regarding the recognition accuracy: I have been able to obtain very good results with OCRKit provided that the quality of the scanned image is high. But I've also had some cases like you mentioned, i.e. errors which I thought the dictionary function should have caught easily.

So overall I am quite impressed with OCRKit, but improvements are of cause welcome. I hope the developers continue to bring their app forward, it has a lot of potential.

Aeschylus Member IconReview+158

I downloaded the demo and ran it on some very difficult documents - poorly made scans, documents with lots of wild fonts, etc. It performed relatively well. I ran it on a PDF i created by removing the text from commercially produced PDF (I exported it to .png, then exported again to PDF in Preview) and it was flawless. It was surprisingly quick on my 5 year old Mac Mini. The verdict - well worth the price. When I saw the PDFs it produces, they are SO much nicer than the CRAZY MS Word docs produced by Adobe Acrobat (they consist of zillions of text boxes and other strange items, not straightforward text and pictures). All in all, what does more for anywhere near this price? Acrobat has a few tricks this doesn't do (for example, editing PDFs, but with great difficulty), but OCRKit is so much nicer to use.

Reply1 reply
Version 1.13

I forgot to say, of course I bought it.

davert Member IconReview+10

The program works at about the same speed as Acrobat but is far easier to use. I was hoping it would not affect the images within my scanned PDF files, but it appears to have degraded them more than Acrobat -- not a huge deal for some purposes, but it makes me wonder why I can't make a PDF file searchable without changing the images in it.

Reply4 replies
Version 1.12

Why doesn't OCRKit start by creating a copy of the file and then working with that? Of course, the user could create a copy and put the copy into OCRKit...


I make copies regardless of the program for the simple reason that OCR increases the size gigantically. And sometimes I just want the darn PDF for my iPhone or iPad and I really don't want a huge file eating up my precious space. So making copies is just a good standard practice, especially since the file at that point is so tiny relative to the final OCR'd output...


@ExactCode -- thanks. I look forward to seeing the update -- generally if I want a PDF to be searchable-ized, it's all images.


@ExactCODE Even if this feature will only work for the simple case of scanned, image-only PDF documents, it would really be great if this updated version you are working on could finally keep CCITT Group 4 compressed images 1:1. I have a big lot of scanned documents lying around waiting to be OCR'd, but there is still no OCR product on the Mac which gets this important detail right (aside from Adobe Acrobat).

Background Information: CCITT T.6 aka Group 4 compression is a very space efficient lossless compression algorithm which was originally developed for facsimile transmission of black-and-white images over telephone lines and data networks. It is optimized for documents with black letters on white background, where it typically yields output files which are multiple orders of magnitude smaller than the results of any other image compression algorithm. CCITT Group 4 has been part of the PDF standard right from the start and so it is still the best choice when storing scanned bi-level images in a TIFF or PDF file.

However, because almost all OCR solutions available on OS X use Apple's PDF Kit framework, and because PDF Kit can only read this format, but does not write it, the resulting files are either much bigger than the original or suffer from a significant loss of image quality as the OCR apps throw away image detail to save space. :-(((

jefferis Member IconReview+1

I just tried the 14 day demo, but the program is acting weird. When it scanned a pdf document, all it did was produce really strange text characters and showed no OCR results:

’1ì*—Á Ö-bîe+ PýæàfŽ§V~ÎM=nTø+ÐxÎX¤äˆ]\°äVoŠ¥ð”–´·V³]ÎKmÏݯðšQqãûë˜Üì‘$8ÆG&¸ÿØ\GâIÞgË3ýÓҋ?z¬åÝ}ðüeoÎcVÞ~@;fº¥‘„B}k–ø}‰ü±cˁ“ƒÖº¥LX;R;òø§Oqç̉†ä‰A‘Ãô °ÎNp)ŎуÅ{°C©8µ$k¿°@;.!‡p®Ñò!Âdç©=ªYHS—S´ÓbîzQp

Reply1 reply
Version 1.12

Prizmo 2 can perform OCR on screenshots too. And it works! You should give it a try.

Yohan199 Member IconReview+0

This app works great! It is super fast when scanning big documents. It also allows for text recognition within the images of my file which is a great feature!

Reply0 replies
Version 1.10.4
Garyrob Member IconReview+3

So far, I'm very pleased. I had been using PDFPen, but it repeatedly crashes or freezes. In my limited experience with OCRKit it has performed flawlessly (on a 22-page document that consistently made PDFPen crash).

Reply0 replies
Version 1.9.4
Brian Duffy Member IconReview-2
Brian Duffy

Ah...I didn't need to give my email and this is not a demo.

I just bought a Mac running Lion and of course my older, expensive, Omni Page OCR software was no longer supported due to Apple asinine decision to drop Rosetta emulation. So, I had to scramble to find a quick and cheap alternative and wow, this did the job! For Free! I also downloaded OCRTools from the app store for $4.99 and than worked well also, maybe just sightly better.

Reply2 replies
Version 1.9.3

By free, do you mean the 14-day trial period?


due to Omni Page's asinine decision not to update old technology.

Fixed it for you.

Adrian Member IconComment+74

What? You cannot download the demo without giving them your email address? Lame.

No thanks, there're plenty of other choices.

Reply1 reply
Version 1.5

"there're plenty of other choices" : I really doubt it. If there's one category lacking in Mac OS software, it's professional OCR tools.

user icon+0
Version 1.13
user icon+1
Version 1.9.6
> 4 11


Current Version (2.x)


Downloads 24,118
Version Downloads 185
License Demo
Date 01 Jul 2014
Platform Intel 64 / OS X / Intel 32
Price $68.20