picatext for Mac

I'm still using this app after 8 years since I bought it. Super convenient and works well. Sure it might not be as accurate as other more expensive products but for a few measly dollars I've gotten 10x the value out of it.

PicaText works "pretty well", which requires some extra explanation. It isn't as accurate as a Google-based OCR, but the main advantage is that *nothing leaves your computer*, which makes it a viable product to use when dealing with text that isn't compatible with a cloud networking solution, or when you don't have a network connection at all. Places where it will often fail on standard screen fonts are 'm' and 'rn' distinctions (when I clip the Participants list in a Zoom window, my name is often "Torn" instead of "Tom"), adding/removing '.' and spaces, and smaller text. These days to improve recognition of those situations I'd use a little bit of ML and user correction feedback (ie, app scans as "Torn" and I correct it to "Tom" then the pattern it scanned as "rn" gets added as a "less likely rn, more likely m" lesson and over time it improves on the fonts people actually use). That said, ML is really most practical on a cloud system where it is learning from everyone in the world instead of just the corrections *I* make, which is essenially why Google OCR is so darned good. Still, PicaText could improve for any one user over time (and perhaps allow an opt-in to upload the corpus of learned patterns to be broadcast out universally on the next version update) by incorporating a locally-run ML mechanism. ML definitely can be done on a laptop (heck, phones do ML pretty well); the downside is not as much data being fed into it as Google gets. I have seen some improvement by taking a MacOS screen shot first (command-shift-4, then click the little screenshot icon in the corner of the screen so it is enlarged mid-screen, then use PicaText on that). It is funny, because there shouldn't be any increase in data doing that, but it picks up text sometimes that way that is missed when just grabbing directly off the screen. Overall, this is an indispensable tool for me and my workflows, well worth the $4 price. It could definitely use some OCR improvements, a feedback / OCR-correction mechanism (obviously only if it is going to "learn" from said corrections), and an option to send off to Google for non-sensitive texts (when connected to an external-facing network). So, "works pretty well" is where I put it. Definitely not the best OCR engine out there, but with the caveat of working locally-only, right now, it seems to be the best option for Mac.

Incredible

This is a very simple and convenient application with the presence of a clipboard for copied text. Does not require online connection for the recognition of symbols. Very lightweight and does not load Mac OS resources. To ensure the accuracy of the copy, you can select the language of the copied text. Allows you to copy message text from open program windows and protected web pages (when you can not do this in the usual way). Copying text is very great (without losing copied characters)! Such applications are quite a lot for Windows OS, but for some reason in Mac OS this useful application is single (the higher its value)! If the developers plan to develop this application, then you might want to implement the following options: 1) ensure that the clipboard is saved when the application is closed; 2) the introduction of a google-translator (but will need to work online) to transfer the copied text to the clipboard; 3) Ability to fix the program window on top of other windows. And the ability to move the program window. 4) The ability to restart the application when it drops. (There were observed periodic fall of the application in the passive mode).

Super-simple, and 100% effective. I had a bunch of scanned pages that I needed to extract sections of text from. Using picatext's keyboard shortcut, it's solid OCR, and it having it automatically place the converted text onto the clipboard ready for pasting, it couldn't have been easier to do. You'll not want to use this application for mass-converting dozens of scanned pages, but for obtaining OCR'd snippets from the odd page, picatext is ideal.