TinyML Talks - Dr. Dominic Binks- TinyML Doesn’t Need Big Data, It Needs Great Data

Explore the importance of high-quality data in tinyML applications through this 34-minute tinyML Talks webcast featuring Dr. Dominic Binks from Audio Analytic. Delve into the challenges of obtaining and processing audio data for sound recognition tasks, learning about effective data sources, gathering techniques, complex labeling strategies, and performance evaluation methods. Gain insights on why great data, rather than big data, is crucial for tinyML success, especially when dealing with edge devices across various applications. Discover how to overcome data-related obstacles and optimize tinyML models for audio recognition and other domains.

Introduction
Context
tinyML
Data matters
Deep understanding of the data
Observations
Data Source
Anomaly Detection
Conclusion