This repository documents various AI/ML projects I have worked on, showcasing implementations of machine learning algorithms and deep learning architectures in Python.
- Abstract:
LipNet is a deep learning model designed for lipreading, enabling the recognition of spoken words from silent videos of a person's lips. This project implements LipNet using a combination of convolutional neural networks (CNNs) for spatial feature extraction and recurrent neural networks (RNNs) with Connectionist Temporal Classification (CTC) loss for sequence learning. It showcases the application of computer vision and natural language processing for real-time speech recognition without audio input. The project is implemented based off this paper.
- Abstract:
This project applies a Transformer-based model to perform sentiment analysis on the IMDb movie review dataset. The goal is to classify movie reviews as positive or negative using techniques like tokenization, vocabulary creation, and padding for input preparation. The Transformer architecture leverages multi-head attention and positional encodings to understand the context of words within a sequence, achieving efficient and scalable sentiment classification.
Feel free to contact me if you have any relevant queries or spot any mistakes in my implementation.