Inspiration]
The inspiration for WHEEZ-AI came from realizing that while therapy is incredibly helpful, it isn't available 24/7. We noticed that many people—ourselves included—struggle most during sudden "crisis moments" or panic attacks when no one is around to help.
We learned that breathing is the only physiological function we can consciously control to directly hack our nervous system and lower cortisol levels. However, during a panic attack, it is almost impossible to remember how to breathe correctly. We wanted to build a bridge between the therapist's office and daily life—an intelligent companion that doesn't just tell you to relax, but actively paces you through it when you need it most.
What it does
WHEEZ-AI is an intelligent biofeedback system designed to de-escalate high-stress and panic situations.
Real-time Assistance: It guides users through scientifically proven breathing techniques (like 4-7-8 breathing or Box Breathing) using visual and auditory cues.
Panic Detection (Optional): It can detect elevated heart rates or irregular breathing patterns to suggest immediate intervention.
Therapeutic Bridge: It logs these sessions, allowing users to track their anxiety triggers over time and, if they choose, share this data with their therapist to improve future treatment.
How we built it
We built the system using Kotlind, the Api of Presage and the Api of gemini
Challenges we ran into
Adpapt the Presage Tehcnologies whith the SDK
Accomplishments that we're proud of
The "Calm Down" Effect: We successfully created a functional feedback loop where the system detects a high heart rate, initiates a breathing exercise, and we can actually see the user's heart rate drop in real-time on our dashboard.
What's next for WHEEZ-AI
Wearable and Hardware Integration: Moving from a stationary prototype to a smartwatch app or a dedicated wearable so users can have support anywhere.
Built With
- elevenlabs
- gemini
- kotlin
- presage
Log in or sign up for Devpost to join the conversation.