Search icon CANCEL
Subscription
0
Cart icon
Your Cart (0 item)
Close icon
You have no products in your basket yet
Arrow left icon
Explore Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Free Learning
Arrow right icon
Arrow up icon
GO TO TOP
TinyML Cookbook

You're reading from   TinyML Cookbook Combine artificial intelligence and ultra-low-power embedded devices to make the world smarter

Arrow left icon
Product type Paperback
Published in Apr 2022
Publisher Packt
ISBN-13 9781801814973
Length 344 pages
Edition 1st Edition
Tools
Arrow right icon
Author (1):
Arrow left icon
Gian Marco Iodice Gian Marco Iodice
Author Profile Icon Gian Marco Iodice
Gian Marco Iodice
Arrow right icon
View More author details
Toc

Table of Contents (10) Chapters Close

Preface 1. Chapter 1: Getting Started with TinyML 2. Chapter 2: Prototyping with Microcontrollers FREE CHAPTER 3. Chapter 3: Building a Weather Station with TensorFlow Lite for Microcontrollers 4. Chapter 4: Voice Controlling LEDs with Edge Impulse 5. Chapter 5: Indoor Scene Classification with TensorFlow Lite for Microcontrollers and the Arduino Nano 6. Chapter 6: Building a Gesture-Based Interface for YouTube Playback 7. Chapter 7: Running a Tiny CIFAR-10 Model on a Virtual Platform with the Zephyr OS 8. Chapter 8: Toward the Next TinyML Generation with microNPU 9. Other Books You May Enjoy

Acquiring audio data with a smartphone

As for all ML problems, data acquisition is the first step to take, and Edge Impulse offers several ways to do this directly from the web browser.

In this recipe, we will learn how to acquire audio samples using a mobile phone.

Getting ready

Acquiring audio samples with a smartphone is the most straightforward data acquisition approach offered by Edge Impulse because it only requires a phone (Android phone or Apple iPhone) with internet connectivity.

However, how many samples do we need to train the model?

Collecting audio samples for KWS

The number of samples depends entirely on the nature of the problem—therefore, no appraoch fits all. For a situation such as this, 50 samples for each class could be sufficient to get a basic model. However, 100 or more are generally recommended to get better results. We want to give you complete freedom on this choice. However, remember to get an equal number of samples for each class...

lock icon The rest of the chapter is locked
Register for a free Packt account to unlock a world of extra content!
A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.
Unlock this book and the full library FREE for 7 days
Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of
Renews at $19.99/month. Cancel anytime
Banner background image