Multi-modal classifications of digits with image and audio modality. One shot learning with Siamese network is used to predict if the given input image-audio pair belongs to same class or not.

Multi-angle Lip Multimodal Video Data

Audio-image classification of emotions

It's a fantastic tool for visualizing medical heartbeat audio data in a unique and engaging way, using NumPy and Matplotlib packages. The script allows you to easily create beautiful visualizations of heartbeats from .wav files.