This is a submission for the Gemma 4 Challenge: Build with Gemma 4
What I Built
SafeGuard AI is an AI-powered disaster preparedness assistant built with Gemma 4 that provides real-time emergency guidance, multilingual support, voice interaction, and live weather-based alerts. The system helps users stay prepared during floods, earthquakes, cyclones, and fire accidents through accessible AI-powered safety assistance. Built using Python, Streamlit, Whisper, gTTS, and OpenWeather API, the project focuses on practical and user-friendly emergency support.
Demo
Code
How I Used Gemma 4
SafeGuard AI uses Gemma 4 E4B as the core AI engine for generating real-time disaster preparedness guidance and multilingual emergency responses. I selected the E4B model because it provides fast response generation, lightweight deployment, and practical reasoning suitable for emergency-response applications. Gemma powers multilingual AI assistance, voice-enabled interaction, and structured safety guidance for users during critical situations.
Cover Image: https://drive.google.com/file/d/1ygD3vT_r2oZVptMdysdb50vcypHbuBJs/view?usp=sharing
This project was developed by: K. Varshitha