This project focuses on classifying lung diseases by analyzing respiratory sound recordings. The system processes audio inputs to identify conditions such as COPD, Bronchiolitis, Pneumonia, Upper Respiratory Tract Infection (URTI), and healthy lungs. Various machine learning algorithms, including Convolutional Neural Networks (CNN), Recurrent CNN (RCNN), Long Short-Term Memory networks (LSTM), and Conformer Transformers, are employed to achieve accurate classification. The dataset used for training and testing is sourced from Kaggle and contains labeled respiratory sound samples. The application includes a user interface with modules for registration, login, prediction, and logout, built using HTML, CSS, and JavaScript for the front end, while the back end is developed with Python and the Flask framework. The goal is to provide a reliable and efficient tool for detecting lung diseases using respiratory audio data. This approach combines audio signal processing and deep learning to support early diagnosis and improve health analysis.

Keywords: respiratory sounds, lung disease classification, COPD, Bronchiolitis, Pneumonia, URTI, CNN, LSTM, Conformer Transformers, Flask.