🫀 Heart Sound Classification using PCG Signals

This project uses machine learning to classify heart sounds from PCG (Phonocardiogram) recordings. It can help detect conditions like murmurs or abnormal heart rhythms.

---

📦 Features

upload heart sounds

• Preprocess and convert to spectrograms
• Classify using a deep learning model
• User-friendly web UI (Streamlit)
• Real-time prediction API (FastAPI)

---

📊 Dataset

• Source: PhysioNet 2016 Challenge
• Files: .wav audio files + .txt labels
• Classes: normal, abnormal

---

🚀 How to Run

1. Install dependencies

pip install -r requirements.txt

2. Train the model

Load the model in your project file 

3. Run the web app

streamlit run app.py

4. Run the API

uvicorn main:app --reload

---

📧 Contact

For questions, open a GitHub issue or email: [kartikguptaasg@gmail.com)