Human Emotion Recognition to Recommend Songs

View Polygence scholar page
Project description
The idea is to develop a project that would recognise human emotion and recommend a song based on it. The implementation of this idea would involve a single machine learning model - a human emotion recognition model. This model would recognise and determine the user’s emotion in real-time through sentimental analysis. Following which, a supporting code would use this recognised emotion to recommend a song for the user to listen to. These song recommendations, however, would be limited to the songs present in the dataset being used - meaning the recommendations would be the same everytime the model is run.


Mentor review
My mentor was extremely helpful at times when I was struggling to debug the code. He gave good and detailed feedback, summary, and instructions for the next sessions.