HearSafeAI

• 

  Hear Safe AI

Inspiration

HearSafeAI – Your AI-Powered Hearing Guardian Inspiration

HearSafeAI was inspired by a tragic incident in which a girl reportedly died after being struck by a train while crossing railway tracks with headphones on, unaware of the approaching train. This incident made us realize how dangerous it can be when people become so immersed in their headphones that they miss critical warning sounds in their surroundings.

In today's world, many people use earphones while walking, travelling, studying, or commuting. Although earphones make our lives more convenient, excessive listening volume can harm our hearing, while missing emergency sounds can put our lives at risk.

We wanted to build a solution that helps people enjoy their audio while remaining aware of their surroundings.

What it does

What Is HearSafeAI?

HearSafeAI is an AI-powered smart hearing protection system designed to help users stay safe by monitoring environmental sound levels and detecting important emergency sounds in real time.

Our goal is to make everyday earphone usage safer by providing timely alerts when potentially dangerous noise levels or critical sounds are detected.

Key Features Dangerous Noise-Level Alerts: Notifies users when sound levels exceed a defined safety threshold. AI-Powered Emergency Sound Detection: Identifies important sounds such as approaching trains, train horns, ambulance sirens, vehicle horns, and other potentially critical environmental sounds. Real-Time Notifications: Alerts users when emergency sounds or potentially harmful noise levels are detected. Hearing Health Awareness: Encourages users to adopt safer listening habits and become more aware of their sound exposure. Continuous Sound Monitoring: Designed to monitor environmental sounds automatically, without requiring users to manually trigger detection.

How we built it

We developed HearSafeAI using a modern application architecture that combines a user-friendly frontend, a backend service, and an AI-based sound classification model.

The frontend was built using React, TypeScript, and Vite to provide a mobile-friendly interface. The backend uses Python and FastAPI to handle application functionality and detection-related information.

For sound classification, we explored Google's YAMNet audio event classification model to recognize different environmental sounds. The backend processes detection results and identifies when alerts should be generated based on the detected sound and its risk level.

Our goal is to connect AI-powered sound recognition with a practical safety application that can help users become more aware of potential dangers around them.

Challenges we ran into

One of our main challenges was integrating AI-based sound detection with the application backend and presenting the results in an understandable way. We also worked on distinguishing emergency sounds from ordinary environmental noise and designing a notification system that could communicate potential risks clearly.

Another challenge was making the application accessible and easy to use, especially for people who may not have technical knowledge.

What we learned

Through HearSafeAI, we gained practical experience in frontend development, backend API integration, AI-based sound classification, and application design. We also learned how AI can be applied to real-world safety problems rather than being limited to theoretical applications.

What's next for HearSafeAI

Our vision is to make hearing protection and environmental sound awareness an everyday part of personal safety.

HearSafeAI aims to help users recognize potentially harmful listening conditions and important warning sounds before they become dangerous situations.

HearSafeAI — Listen Smart. Stay Aware. Stay Safe.

Built With

• artificial
• audio
• classification
• fastapi
• healthcare
• intelligence
• learning
• machine
• monitoring
• python
• react
• real-time
• typescript
• vite
• yamnet