Red Rooster: Craveware

Project Overview:

crave_ware is an innovative AI software program developed by whiteGREY that listens to and categorises hunger cues (aka tummy rumbles). This technology not only detects stomach rumblings but also places a customised meal order on behalf of the customer based on the intensity and duration of their tummy rumbles.

Key Features:

• AI-trained model using TensorFlow.js and CNN architecture
• React-based PWA for offline-first experience
• Bluetooth connectivity for seamless data transmission

Tech:

• TensorFlow.js: We used TensorFlow.js to train our AI model on tummy rumbles, enabling it to distinguish between hunger cues and other bodily noises.
• Convolutional Neural Network (CNN): Our AI model uses a CNN architecture to analyse audio data from the in-built microphone on the t-shirt. This allows us to extract meaningful features from the audio signals.
• React: We built a Progressive Web App (PWA) using React to provide a seamless, offline-first experience for users.
• Bluetooth: The t-shirt's microphone uses Bluetooth connectivity to transmit audio data to the user's device.