Inspiration
We love to listen to music while we read to help immerse ourselves in the world, and we wanted to create a platform where users could create custom playlists for their reading sessions.
What it does
Orpheus uses NLTK to analyze the positivity of a book based on a text file, then uses that score to generate a Spotify playlist of 100 classical songs with a corresponding valence score. When users use the Orpheus e-reader to read their book, they can also play the matching playlist using an embedded player.
How we built it
We used Python and NLTK to analyze the sentiment of the text, and then used a Flask app to authorize the system account and add playlists. We also used Figma, HTML, and CSS for the front end.
Challenges we ran into
The Anima design-to-code platform on Figma wasn't giving us the results we wanted, so we had to convert the designs from Figma to CSS by hand.
Accomplishments that we're proud of
Figuring out the Spotify API authorization was tricky, but finally getting it to work was a big accomplishment because we can use it in so many other projects going forward.
What we learned
We used Flask for the first time, set up a website using Github for the first time, set up refreshing authorizations on the Spotify API, and made a website based on a Figma design.
What's next for Orpheus
Letting users add their own books. Right now we have to add books manually, but ideally Orpheus would allow someone to upload a book themselves and save the corresponding playlist to their own Spotify instead of a system account.
Log in or sign up for Devpost to join the conversation.