A Machine Learning-Based App for Sound Detection and Speech Recognition for the Deaf and Hard of Hearing

This project uses AI to improve safety and communication for the deaf and hard-of-hearing community in Saudi Arabia. By combining real-time sound detection and speech recognition, it offers alerts for potential dangers and supports communication. The system is built with Android Studio, TensorFlow Lite, and Speech-to-Text APIs. The core model uses YAMNet for feature extraction and a Deep Neural Network (DNN) for classification, achieving 95.77% accuracy. Future updates will include features like customizable alerts, hazard detection, sign language support, and location tracking, aiming to enhance safety and inclusivity.