This was the main project for my course CS543: Massive Data Storage and Retrieval. The aim of the project was to be able to predict genre of a song with time sequencing. Although, genre recogntion is not a new idea, what makes this different from other such projects is the use of direct musical data and no lyrics based features.
Main features of the project:
The Mel Spectogram for our data.
Using Mel spectograms to represent musical features
Use a Convolutional Recurrent Neural Network which was later replaced by Time-Distributed layers to improve accuracy.
The final model was deployed and visualized on web by using TensorflowJS.