Deep Learning for Multimodal AI

Modern Artificial Intelligence (AI) employs a variety of modalities, including image, audio, and text. Introduction to deep learning and its domain-specific applications from computer vision and Large Language Models (LLMs) to audio and speech recognition. Hands-on experience building deep learning models (CNN, RNN, Transformers, etc.) and AI systems using frameworks such as PyTorch and Tensorflow Keras. Survey of State-Of-The-Art literature on multimodal AI such as CLIP (Language-Image) and CLAP (Language-Audio). Opportunity to carry out a mini AI research project following industry standard practices.