SoundCrowd

Entry Screen
Guest View (Voted)
Guest View
Host View

Inspiration

As all college students know, bad music makes for bad parties. We decided to build an app that would put music choice back in the hands of the partygoers.

What it does

From the party host's perspective, this app allows the user to create a room which can receive input from party guests and can control the host's Spotify based on the guests' decisions. For the guest, joining a room offers the user a selection of four songs to play next, from which they can vote. The majority vote will be played next! These suggestions for up next are generated automatically by the app through machine learning techniques leveraging party voting history and song clustering in order to tailor the suggestions to best match a party's mood. n

How we built it

We developed this app in Swift and made extensive use of the Spotify API along with the Spotify SDK and Firebase SDK for backend development.

Challenges we ran into

The greatest challenge came in timing the various asynchronous calls such that all users both hosts and guests had appropriate access to their desired information at all times.

Accomplishments that we're proud of

Our user interface is clean and simple, which really highlights the ease-of-use design model that we really hoped to achieve. All of us firmly believe that this app can help solve a valid problem, particularly in a college community, so we're just happy to have seen this development all the way through.

What's next for SoundCrowd

Development on SoundCrowd is far from over. There are plenty of refinements to be made of our user interface to ensure that the process is painless. We're hoping to add additional features such as proximity based party detection to obviate the need for party ID's at all. Then, we're hoping to enrich our dataset and techniques for the ML portion of our project to give all partygoers the best possible suggestions to match their moods.