AirSure
Home Page
Scanner
Completed Scan
Boarding Pass
Flight Alerts

Inspiration

Air travel is full of small but costly mistakes—bringing restricted items, missing gate changes, or not hearing important announcements in noisy terminals. We were inspired by how often these issues happen, even to frequent travelers, and saw an opportunity to use on-device AI to make travel smoother, more reliable, and less stressful.

What it does

AirSure is an on-device travel assistant that helps users avoid common travel mistakes in real time. It has two core features:

A carry-on checker that uses your phone’s camera to detect items as you pack and flags potential issues like oversized liquids or restricted items.
An announcement listener that processes nearby airport audio and extracts key updates—such as gate changes, boarding calls, and delays—then displays them clearly on your screen.

How we built it

We combined real-time computer vision and audio processing directly on-device to ensure low latency and privacy. For the carry-on checker, we used object detection models optimized for mobile to recognize common travel items. For the announcement listener, we implemented a speech-to-text pipeline paired with keyword extraction to identify relevant travel updates. The app integrates these systems into a simple, intuitive mobile interface.

Challenges we ran into

One of the biggest challenges was balancing accuracy with performance on-device. Running real-time vision and audio models efficiently without draining battery or requiring cloud processing required careful optimization. We also had to handle noisy, unpredictable airport audio and ensure that only relevant announcements were surfaced to the user.

Accomplishments that we're proud of

We’re proud of building a fully on-device system that works in real time while maintaining user privacy. We successfully demonstrated both visual detection for packing and live audio parsing for announcements, creating a seamless experience that feels genuinely helpful in real travel scenarios.

What we learned

We learned how to optimize machine learning models for edge devices and how important UX is when presenting real-time information. We also gained experience handling messy, real-world inputs like cluttered bags and noisy audio environments.

What's next for Airsure

Next, we want to improve model accuracy and expand item detection to cover more edge cases. We also plan to add flight integration for personalized updates, multilingual announcement support, and smarter notifications based on user context. Ultimately, we aim to make AirSure a comprehensive, reliable travel companion from home to destination.