The size of deep learning networks is decreasing. Very little. A model developed by the Google Assistant team that is just 14 kilobytes large?small enough to run on a microcontroller?can detect words. With the help of this useful book, you'll learn about the topic of TinyML, where embedded systems and deep learning work together to enable amazing things to be done with little gadgets. Pete Warden and Daniel Situnayake describe how to teach miniature models that can be used in any setting. This manual leads you through constructing a number of TinyML projects step-by-step and is ideal for software and hardware engineers who wish to design embedded systems utilising machine learning. No prior knowledge of microcontrollers or machine learning is required.