Inspiration
It can be difficult to figure out what the sentiment of a piece of text is especially for people who aren't fluent in English or for many people who are neurodivergent. We aim to help people who struggle with this by conveying emotions through music instead. In addition, we wanted to create a fun web application that anyone can use to find songs from text.
What it does
It takes input, such as plain text or a text file and uses word similarity scores to analyze the sentiment and find the main emotions in the text.
How we built it
We used HTML/CSS, JavaScript, and Python to build an interactive web app that uses the NLTK WordNet model to score the similarity of words to a list of emotions. We then took those top emotions and associated each of them with a song that became a seed Song for spotify to generate similar songs for our playlist.
Challenges we ran into
Finding the right model to compare the individual words to the emotions, getting a proper domain, implementing the Spotify API, and integrating frontend and backend.
Accomplishments that we're proud of
We are proud of the final result of our project and the playlists it creates as well as some of the technical aspects including being able to filter out articles, punctuation, and miscellaneous words that have no emotion in our emotion finding files
What we learned
How to properly use tuples, NLTK WordNet, more in-depth knowledge of Python, and how to set up a website
What's next for Musing the Music
Allowing users to connect to their account from the front end as opposed to editing the code
Log in or sign up for Devpost to join the conversation.