This website moodifier detects the human facial expressions, using ML model, which we trained using over 16000 pictures on kraggle. So basically a user has to upload a picture of theirs. The model (which is deployed on our main web page using flask) will detect the emotion, and will redirect the person to the respective emotion pages. There, user can find useful resources and playlists of songs that they can listen to. User just need to click on a song and it'll start playing in the same tab itself in a small embedded youtube frame. They can also find resources that might be helpful when an user is feeling sad.


Music is one of the most soothing medicine for all our moods. It enhances productivity, and it's related to our life in a large number of ways, no one can deny that fact. So let's use this powerful weapon to tidy up our moody minds.

What it does

Moodifier is a website that detects the emotion from a newly taken or uploaded photo and redirects them to personalized songs and resources according to their moods.

How we built it

--designing tool: Canva and Microsoft Paint 3D.
--frontend :HTML CSS and JS.
--backend : flask (Machine learning libraries from tensorflow), linode (for hosting our webapp)

Challenges we ran into

We had a tough time trying to organize the amount of data and files that we're integrating. Also, we were not very used to with flask, so we spent a lot of time to figure it out.

Accomplishments that we're proud of

  • Making something helpful for the community
  • Successfully making the website work!
  • Managing to create nice playlists and graphics, for a user-friendly environment

What we learned

  • detect emotions from a photograph, using keras and tensorflow libraries
  • gained more experience in CSS
  • explored so many amazing songs lol
  • gained more experience in webpage design

Sponsor services used

  • Godaddy domain (music-with-usn.co)

What's next for Moodifier

We plan to add more features, including a way for users to create accounts where they can input and save their data for a better personalized experience. We also want to make a mobile app version of Moodifier for more convenient usage.

Built With

Share this project: