Reconstruction of music using EEG signals. Inspired from Brain2Image, reconstruction will be done using generative models like variational autoencoders(VAE) and conditional generative adversarial networks(GANs).
For Data cleaning and preprocessing techniques like MFDXA will be used.
This project is funded by Intel under Intel Early Innovator funding.
- Literature Study (in progress)
- Data Collection (in progress)
- Data Preprocessing
- implementing different models to the dataset
- Data post processing and Analysis of results
- Python 3.6
- Librosa for audio analysis
- PyTorch for Deep Learning models
- MATLAB for preprocessing