Inspiration
I am someone that really likes listening to music and being able to associated memories or even activities with certain songs or emotions. So, being able to create a system that can help me find some songs that fit those moods was really interesting to me.
What it does
The time capsule radio takes in a prompt of whatever memory you want. With that, the AI detects what mood that memory gives off and tries to correlate it as closely as possible to about 15 different moods that we created. From there, it reads out of the Dataset.csv file and chooses songs that fit that memory.
How we built it
I created all of this in Pycharm, I started with the main.py file that just contains some sorting methods. Then, came the app.py file, which was the true frontend of the project. It required the style.css file to be created to add some aspects to the UI and make it so that I could actually have the filters within the system.
Challenges we ran into
I really struggled with figuring out how to really even start. It took a lot of researching to figure out how to even integrate an open source AI system into Pycharm. However, once I figured that out, it took a lot more learning to realize what other systems I needed.
Accomplishments that we're proud of
I am most proud of being able to create the frontend, even though it isn't the most appealing. It was my first time ever creating a frontend and I did it all by myself. I did this entire project by myself, which is another accomplishment I am proud of.
What we learned
I learned a lot about AI pipelines, as well as using streamlit. I only ever really focused on the backend, so I had to learn a lot about how to use css and other tools to create a frontend.
What's next for Time Capsule Radio
I think the next implementation is heavily on the UI of the application, making it look nicer, more polished, maybe even adding a feature to get the playlist onto some other platforms, not just Youtube.
Log in or sign up for Devpost to join the conversation.