RapidAI

Inspiration

Recent destruction during the LA fires increases because of understaffing and long 911 wait times due to the overwhelming of call centers.

What it does

Steps in when there are no staff available to take the call and analyze the situation in terms of emergency level - (minor, major, critical) based on the description given, it also analyzes the fear and anxiety of the situation. Once the call is on the caller dashboard, the call staff can dispatch police immediately to the place, and a message is sent immediately to the police with all the necessary information, including the location and emergency details

How we built it

We used Twilio to automate calls and text messages, which were then sent to the Whisper API for transcription. After transcription, the text was processed by our AI model for categorization. For the map functionality on our dashboard, we integrated the Google Maps API.

The backend was developed using Python and Flask, while the AI model was built using PyTorch and machine learning techniques. For the frontend, we utilized Tailwind CSS to create a responsive and user-friendly interface.

Challenges we ran into

Training the AI to correctly categorize emergency levels based on varying caller descriptions. Ensuring fast and accurate message transmission to emergency responders. Being able to asses sentiments and add that information to the dashboard.

Accomplishments that we're proud of

We are proud of successfully creating an AI-based system for emergency response. Creating a solution to help people in emergency situations in any place Getting the phone calls to work properly