CantoRead Blog

Here’s what happened last week in the world of CantoRead Supporting iOS 14 This was actually quite easy, all I did was lower the version number. This means more people will be able to install CantoRead. Flashcards I created a simple flash card feature. When you process an image, CantoRead creates flash cards for all the words in that image. Let’s say you saw the following image. If you’re me, there will be a bunch of words you recognise on this sign, and a bunch that you do not.

Great Success CantoRead passed review a few hours after I published my last blog post. They really have sped up their review process. Back in the day I would deploy to Android first since they didn’t have a review process. CantoRead is very simple at the moment. All you do is select a photo, and then scroll through the Chinese text in that photo. Here’s a demonstration: Reinforced Learning At

Thursday was a good day, I was in a “get that shit done and out the door” sort of mood. Originally I was planning to have the following features in CantoRead: Extract Chinese text from user provided images Show an advertisement while processing a user’s image Allow users to buy a monthly/yearly subscription to turn off ads On Thursday I decided to cut feature #3, Apple’s payment APIs are quite confusing to work with.

Elevator Pitch CantoRead is a mobile application that allows you to learn Cantonese through the Chinese characters you’re exposed to every day. Busted Elevator Pitch Imagine you and I are stuck in a busted elevator. We’ve been in there for quite some time, and you say to me “Tell me a story Angus Cheng”. “What sort of story do you want to hear?” “Tell me about that new mobile application you’re making”

I’ve tried out a few different Optical Character Recognition (OCR) services for Canto Read. First I tried out Apple’s Vision Framework, it all happens on the device which is nice, however the results I got weren’t very good. It failed to detect a lot of text, and the text that was recognised often was incorrect. So I thought, “Ahh that’s a bummer” and moved on to using Google’s similarly named