Projects

Developed an advanced deep learning model in TensorFlow utilizing LSTM architecture to generate piano compositions, trained on a diverse dataset of over 5,000 classical pieces. Enhanced note transitions and harmonies by integrating sequence attention mechanisms for improved realism. Currently expanding the project to support multi-instrument compositions by leveraging Transformer models and multi-track datasets, enabling the generation of dynamic and richly orchestrated musical pieces.

Developed a real-time hand gesture recognition system leveraging MediaPipe, TensorFlow, and OpenCV. The system detects 21 hand keypoints and classifies 10 distinct gestures using a pre-trained neural network, achieving 95% accuracy on a custom dataset. Designed for applications such as VR control, sign language translation, and music creation, this project highlights my expertise in computer vision, deep learning, and real-time interactive systems.

Designed a Sentiment Analysis system leveraging advanced Natural Language Processing (NLP) techniques to classify text as positive, negative, or neutral, extracting valuable insights from sources like social media, customer reviews, and feedback. Utilizing machine learning and deep learning models such as TF-IDF, word embeddings, LSTMs, and transformers, this project demonstrates expertise in text analytics, AI-driven decision-making, and real- world applications like brand sentiment tracking and automated content moderation.