QuickMath was developed as part of a fourth year module named Gesture Based UI Development. The idea was to create an educational game where users were able to solve Addition/Subtraction/Multiplication math equations with three different difficulties. The the math equations are dynamically generated based on game-mode and difficulty selected and the user uses their phone's microphone to answer the equations.
Successfully solving a mathematical equation rewards the user with a score point for their selected game - different difficulties provide more points per solve. The user can register/login using one input of a Username and a picture of their face.
The Android application was written using Ionic 3 with Native Camera and Native Speech assisted with Google Assist for speech recognition.
In this repository you will find the source code and APK for the application.
The architecture of the application on the front end is based on Angular4 Components and Providers. There are two components: one for the authentication and the other for the game.
The providers supply the application with service-like functions. There are currently 3 providers: AuthProvider, ScoreProvider and SpeechProvider.
The AuthProvider supplies the login component with HTTP functions to contact the back-end for logging/registering a user.
The ScoreProvider supplies the game component with HTTP functions to contact the back-end for updating highscores.
Finally the SpeechProvider is used to supply the Game component with functions for acquiring permissions to use the microphone.
Using the native camera on your android device, you can take a photo ID that acts as your password to the username. This enables the use of a facial gesture to authenticate users. We chose to do this gesture because it seems like this is a field of authentication that is being developed rapidly. It also acts as a fun-feature for kids to authenticate themselves with the help of their parents.
Using the native microphone on your android device and google assist Speech Recognition, you can record your answers to the math sums with a voice gesture. This allows users steer clear of the usual input boxes, which leads to keyboard popups, which hide the screen, which makes the game harder to play and so and so and it doesn't feel natural. You could argue to have cards with answers but that is basically an MCQ-type game where the answer is given to the user and this takes away the difficulty aspect. Voice was perfect!
Overall the project was very fun to do and it was great learning about recognition technologies. If we were to do it again, we might choose a different framework for the front-end as Ionic was giving us the biggest issues when trying to set up an environment where both team mates could debug the application.