The inspiration for this application is to break down music into information easier for users to understand and possibly for their own use.

What it does

Hubbub is an instrument detecting web application. When given an audio file, it can break down the instruments used and return it back to the user.

How we built it

Hubbub was built using Django for the servers, firebase to store the audio files. We used python and machine learning to train the data so our application will be able to detect the instrument used in the audio file given. Html, css, javascript, and react were used for the frontend.

Challenges we ran into

We spent a large portion of our time trying to setup the Django server. Another problem that we faced was finding datasets for different instrument sounds.

Accomplishments that we're proud of

One of the accomplishments we are most proud of is being able to complete the project within 24 hours. In order to do so, we did code sprints for each feature before taking a proper break.

What we learned

We learned how to manage time more efficiently with a smaller group size from this hackathon. We also gained a stronger understanding of machine learning and react.

What's next for Hubbub

Hubbub's next step will be being able to detect the notes used in addition to the instrument in the future. We will also implement useful features such as beats per minute and possibly detect scales or chords to further help music students or students interested in dissecting the fine details of music that the natural human ear could not decipher.

Share this project: