This project aims to develop a deep learning approach for music recognition. The goal is to create a system that can accurately identify songs or music tracks based on their audio features. The project is in its initial stages and currently explores various deep learning models, including ResNet50, LSTM, and CNN with n-fold cross-validation.
git clone https://github.com/your-username/music-recognizer.git
The project contains Jupyter notebooks with initial implementations of the models:
basic_CNN.ipynb: A basic Convolutional Neural Network (CNN) approach.LSTM.ipynb: An implementation using Long Short-Term Memory (LSTM) networks.resnet50.ipynb: An implementation using the ResNet50 model.
To run the notebooks, open them in Jupyter Lab or Jupyter Notebook and execute the cells sequentially.
https://github.com/mdeff/fma/tree/master -> fma_small.zip
Contributions to the project are welcome. If you have ideas for improvements or new features, please open an issue or submit a pull request.
Vikas Kumar -> https://github.com/thisisvk45
Rachit Gupta -> https://github.com/rachitgupta007