Imagine this, you are trying to find a nice playlist to workout, study or even meditate to. However, being your indecisive self, you simply cannot commit to one playlist for a long duration of time. We have all been in a situation like this before and although the time wasted on this dilemma may seem minuscule, it can accumulate to hours of wasted time and ultimately, opportunity. From this idea, Bella flourished.
What it does
Bella is a mobile application that leverages the existing technology of Interaxon's Muse headband combined with proprietary machine-learning algorithms to curate a dynamic Spotify playlist perfectly suited for the user's complex and unique state of mind.
How we built it
The machine learning algorithm was built with Python and the integration of Muse was done with Python. Finally, the whole program was packaged together and set up on iOS through Swift.
Challenges we ran into
The main challenges that our team ran into was with integrating our neural network with the Muse data and Spotify's API. The Muse data comes in 5 different "wave-length" values and the challenge came from translating the accumulation of the values into distinct "moods." Also, complications arose from Spotify API's authentication process and its limited/convoluted documentation. There also difficulties using multi-threading techniques in objective-c to build the integrated GUI.
Accomplishments that I'm proud of
Overall, our group was extremely pleased that we were able to successfully integrate Muse, an innovative piece of technology, with machine learning. Since they are advanced concepts, we were honestly extremely pleased to be able to grasp and implement both concepts in such a short amount of time.
What we learned
We learned how to successfully integrate the Spotify API along with making a neural network from scratch. Overall, as a group we learned how to successfully integrate Python, Java and Swift to create a mobile app that works harmoniously.
What's next for Bella
We believe that the next step for Bella would be to integrate the software and technology into a smaller footprint. Currently, the software relies on a cumbersome headset; the dream would be for Bella to be able to work inside headphones and ultimately become a normal part of music consumption.