Goti is an advanced voice assistant powered by the LLaMA 3 model, featuring real-time voice-to-text conversion and custom fine-tuning. Built with Python and cutting-edge speech recognition technology, Goti blends trendy and professional tones to offer engaging and precise interactions.
ScrollAssistant is an innovative website that enhances news reading by enabling users to control scrolling, opening, and closing of news articles through voice commands. Built with React.js and powered by Alan AI, it also features an integrated chatbot for an interactive user experience
deploys a FastAPI app using Hugging Face Transformers for text generation within a Docker container. This setup simplifies deployment on platforms like Hugging Face Spaces, with key files including app.py for the application code, Dockerfile for the image, and requirements.txt for dependencies.
This project is a real-time chat app built with Jetpack Compose for the UI and Firebase for backend services. Features include real-time messaging, user authentication (email/password and Google Sign-In), status image sharing, profile management, password reset, and push notifications.
.9aaeb6ef.png&w=3840&q=75)
This Image Processing Application allows users to apply various filters and transformations to grayscale images. It is built using Streamlit, a popular Python framework for creating interactive web applications. Users can upload a subject image, choose from several filters, and view the results directly in the application interface.
.5c319e66.png&w=3840&q=75)
This computer vision project uses AI to enable virtual mouse control and hand gesture-based volume adjustment. It translates hand movements into mouse actions and adjusts volume based on specific gestures, providing an intuitive and hands-free interface.
Led the development of WristMinder Pro, an automated radiological report generation tool focused on pediatric wrist anomalies. Utilized the GRAZPEDWRI-DX dataset and the IDEFICS visual language model, achieving strong evaluation metrics like ROUGE and BLEU. Collaborated with radiologists and experts, contributing to a robust dataset and creating an Android app for streamlined X-ray uploads