Explore a conference talk from tinyML Asia 2021 focusing on airborne sound maintenance in remote sites using low-power federated learning. Delve into the business and technical rationale behind selecting TinyML for Contextualized Airborne Sound in Predictive Maintenance. Discover how this solution aims to minimize operational downtime, reduce working capital for spare parts, and lower retrofitting expenses for existing machine infrastructure. Learn about the innovative approach using low-power sensors and federated learning to continuously improve the model while adhering to GDPR regulations. Gain insights into real-world industry examples, the concept behind the solution, and a demonstration of its implementation. Understand the background, vision, and edge processing involved in this cutting-edge application of tinyML technology.
Airborne Sound Maintenance in Remote Sites Using Low Power Federated Learning
Overview
Syllabus
Introduction
Why do companies see this as important
The problem
The concept
Examples from industry
Federated learning
Demonstration
Background
Vision
Edge Processing
Sponsors
Taught by
tinyML
