Myuzic
I was bench pressing weights at the gym, one hot summer afternoon when my phone started playing Bridge Over Troubled Water by Simon and Garfunkel. Unfortunately, I was in the middle of an intense set and I could not just change the song. This led me to search for a music player that is able to pick songs according to my tastes and preferences. After not finding an app that I liked (my data plan is too small for Spotify), I decided to create my own.
How It All Works
Myuzic has 3 major components: the App itself, a back end, and an API. The app uses the API to transfer data about the user's listening habits to the back end. The back end uses artificial intelligence to compile a list of songs that the user would most likely listen to. The back end is written in Python and uses a Keras and Tensorflow sequential linear regression model in order to predict songs with a loss of approximately 0.117.The Myuzic IOS application is written in Swift with the backend API being written with JavaScript, Node.JS, Express, and MongoDB for data storage. The source code for the app, API, and backend is provided below.
Your Music Library Like Never Before
Clean and Interactive UI
The App
The API
The Back End