Fantasmix
A Project by Stefano Arnone, Jakob Au, Mohammed Hassan, Tyler Hooper, and Mohamed Mohamed
Our goal is to create a free website with an algorithm that generates instrumental compositions based off of inputted user settings such as genre, instruments/sounds, and songs/bands for the public. This algorithm guesses user’s preference and creates a 2 - 3 minute song based on their inputted settings. There are other similar current projects such as Google’s Magenta, DeepJazz, BachBot, and more, and we aim to learn from them and create our own version. We hope to use our fresh young minds to find a unique solution and contribute to the creation of new, computer generated, music.
Our motivation to make this project comes from our mutual interest in music, both listening and creating. This program will help users find new tastes in music and new songs to listen to, as well as inspiration for creating new music. Users could take instrumentals created on the site and develop them on their own. We see our app as a starting point for people getting into music development. Users will be able to find their likes and dislikes in music and allow themselves to create/ personalize music.
Road blocks include creating something unique from the many other researchers working on the same project for years (major roadblock), creating a complex algorithm dealing with artificial intelligence, and integrating multiple ideas into a single project. People have been working on this for years, and a couple college freshman might not be the winning factor. Creating something different in our own way is yet to come. We see roadblocks ahead when it comes to the randomization, so very much research will have to be done. Also, when it comes to finding and storing the music files, we will have to figure out how we will implement them into the program/site.
Our order of major tasks is to create a spreadsheet of already made algorithms, to conduct individual research, to create our version of the music generating algorithm, and finally to design and implement the user friendly website. Researching algorithms will definitely take some time, as we have to figure out how the main feature of the program can be put into code. We will also have to do some research on databasing many music files and linking them to our corresponding website/program.
Through this process, we envision ending up with a final website that houses the program we are creating. Ideally the app will be integrated into the website, including most of the features we are outlining to create. Our first major milestone will be to have one working genre of music implemented, including instruments and melodies that can be combined and exported into a file that users can download. Upon reaching this goal, we will work on adding more to the project.
TEAM PROFILE
Scrum Master / Technical Designer - Jakob Au
Operations Manager - Mohammed Hassan
Product Owner / Technical Designer - Stefano Arnone
Business Manager / Front-end Developer - Tyler Hooper
Aesthetic Developer / Front-end Developer - Mohamed Mohamed
Static Site Github:
https://github.com/thooper159/thooper159.github.io
or
fantasmix.net