Inspiration

Our inspiration came from a real-life situation where someone in danger couldn't get help in time. This made us realize how many emergencies go unnoticed in public spaces. We wanted to create a device that could "hear" these situations and respond instantly, even if no one else is around.

What it does

AlertEars is a smart device that listens for sounds like screams or cries for help in real time. When it hears something serious, it alerts the police or other systems. It’s waterproof, protects privacy, and runs all day in public places like subways or streets.

How we built it

We made AlertEars with a sound detection module linked to a microcontroller. It uses trained AI models to spot emergency sounds. The device connects to the cloud to alert the police. We also built a tough case so it works well outside.

Challenges we ran into

Training the sound detection model to accurately recognize real distress sounds while ignoring background noise was difficult. We also had to balance privacy concerns, making sure it only records when necessary. Hardware integration and weatherproofing were also tough.

Accomplishments that we're proud of

We successfully built a working prototype that detects emergency sounds in real time. It can function in noisy environments and sends alerts within seconds. We’re also proud that the design is small, low-cost, and suitable for real-world use.

What we learned

We learned a lot about machine learning for audio, hardware-software integration, and privacy-focused design. Most importantly, we learned how powerful simple tech can be when applied to real-world problems.

What's next for AlertEars

We plan to improve sound recognition accuracy, test in more real-life environments, and partner with local governments or safety organizations. We’re also exploring adding multilingual voice alerts and GPS tracking for faster emergency response.

Built With

Share this project:

Updates