A Covid Detector App illustrating Android development practices with Android Jetpack.
The main purpose of this application is to identify if user has infected with Covid-19. With this application we can detect if a person gets infected by covid-19 using cough sound. Hence, we can anticipate the infection of covid-19 before it gets worse.
Android Covid Detector is currently under development. Some changes will be updated (such as database schema modifications).
We developed our app using Android Jetpack guideline. Android Jetpack is a set of components, tools and guidance to make great Android apps. They bring together the existing Support Library and Architecture Components and arrange them into four categories:
Android Covid Detector demonstrates utilizing these components to create a simple detect covid using sound of cough.
- master (Added main feature: Covid Detection with tensorflow lite)
- development-1.0 (Added main feature: Covid Detection with tensorflow lite)
This project uses the Gradle build system. To build this project, use the
gradlew build
command or use "Import Project" in Android Studio.
There are two Gradle tasks for testing the project:
connectedAndroidTest
- for running Espresso on a connected devicetest
- for running unit tests
For more resources on learning Android development, visit the Developer Guides at developer.android.com.
- Foundation
- AppCompat
- Android KTX
- Test
- Architecture
- Data Binding
- Lifecycles
- LiveData
- ViewModel
- Navigation
- Room (Local Database)
- Firebase real-time database (Remote Database)
- Firebase Storage (Remote Storage)
- UI
- Fragment
- Layout
- Third party and miscellaneous libraries
- Glide for image loading
- CircleImage
- dmax Spot Dialog
- TensorFlow Lite (Machine Learning)
Updates will include news feature using API to get latest information about COVID-19 for users.
- Download Dataset
- Preprocess Dataset
- Make model and training dataset
- Create Prediction
- Result Prediction
- Convert keras model to tflite model
- Create VM Instance for running Jupyter Notebooks, Tensorflow, Keras.
- Create firebase to store user data
- Create custom model tflite on firebase