Inspiration
Mental health issues, mood swings, depression, and anxiety are some of the most common psychological problems. And our goal is to help the users to become mentally balanced.
What it does
Our android application receives and processes speech and text from the user and performs psychological analysis. Based on the current mood, we recommend musics and videos that will help balancing the mood.
How I built it
Firstly, our android uses speech to text API to parse the user speech input. Then, we make a REST API call to the Alchemy Language API from IBM Watson for tone analysis and emotion analysis. Using the emotion components, we select a music and a video from the database constructed using scikit-learn with linear regression model.
Challenges I ran into
We didn't have enough time to have prototypes for other ideas to improve our product.
Accomplishments that I'm proud of
Successfully built the fully functional minimal viable product.
What I learned
We learned about teamwork and integrating various APIs in our product.
What's next for Balance
Improvements:
- As of currently, our database is too small for recommendation. Using larger database of music and video.
- Trying to use hierarchical model rather than linear model for larger datasets.
Log in or sign up for Devpost to join the conversation.