Audio MNIST classifier implemented in Pytorch Lightning Using a CNN to classify recordings of digits (0-9). current acc - 0.98 dataset: https://www.kaggle.com/datasets/sripaadsrinivasan/audio-mnist?resource=download TBD: add augmented data to the dataset to improve accuracy. create an inference class/package.