Just released our 3.2.12 update on iOS, which includes an early access version (optional - enable in settings) of our revamped OCR system; not all of the changes are in yet (still the old UI) but we've made some big low-level changes, including an awesome new motion detection algorithm that makes a huge difference in the usability of live OCR. Along with that public update, we also quietly added a brand new OCR recognition engine we're testing, but we aren't offering that part to everyone yet since we want to make sure it works well before we start taking people's money for it (we have to pay new royalties on it). Eventual plan is to let people alternate between both engines as we think each offers some benefits in certain scenarios. So we'd like to recruit some beta testers to try that out; if you've got an iPhone/iPad and are interested (and preferably have sent us beta feedback before), PM or email and we'll get you added to the beta. This is all coming to Android too, incidentally; in fact, one of those internal changes was moving a lot of our OCR code from platform-specific Objective-C / Java to cross-platform C++ so that we can push future OCR algorithm refinements rapidly to both platforms. It's just much easier to do new OCR development on iOS; much better native code support plus a far far smaller range of camera drivers / hardware to worry about.