Inspiration
All three of us love music and wanted to create something that would speed up the process of finding songs that either reflect or change our current emotional state.
What it does
This program takes pictures of the user using real-time video capture and evaluates the current emotion on the user's face. Based on the result, a playlist will be added to your Spotify account.
How we built it
To build this, we had to first implement a classification AI with deep learning that can read the emotions on users' faces. We then connected the AI to the Spotify API to pick songs to reflect or change the resulting mood from the picture. We then tied the two parts with a simple UI to complete the project.
Challenges we ran into
The biggest challenges we ran into were how to implement the AI because we didn't know how to make one. Another challenge was how to access Spotify accounts to add playlists. The last challenge was using a UI to combine both without getting access_token errors.
Accomplishments that we're proud of
We are unbelievably proud that we were able to finish the beta version of our project. It took a lot of work, not just coding and debugging, but also reading and understanding documentation.
What we learned
The biggest thing we learned was the fact that we really can learn anything in a short time if we really grind it out. Also, this project made our python programming even better than before.
What's next for Musical Recognition
We would love to improve our AI to make the recognition percentage as high as possible along with understanding access tokens for the Spotify API. We would also like to make the UI nicer and better for users.
Log in or sign up for Devpost to join the conversation.