Somokolon LabsSomokolon Labs
AI Products
Prototype

ResoNet

On-device respiratory screening from cough audio

A privacy-preserving respiratory-disease screening app that runs entirely on-device. Audio preprocessing and a ResNet model were reimplemented in pure Dart with ONNX Runtime, cutting the app footprint from 600 MB+ to roughly 70 MB with no loss in accuracy.

Air quality monitor shows levels of pollutants.

Key features

What ResoNet does

Fully on-device inference

Pure-Dart audio preprocessing

~70 MB footprint, down from 600 MB+

Works offline for privacy

Built with

ONNXResNetDartEdge AI

Status

Prototype

Request a demo

Have a problem worth solving with software?

Tell us what you're building. We'll help you scope it, build it, and ship it.

Start a project