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
Data Augmentation with Python

You're reading from   Data Augmentation with Python Enhance deep learning accuracy with data augmentation methods for image, text, audio, and tabular data

Arrow left icon
Product type Paperback
Published in Apr 2023
Publisher Packt
ISBN-13 9781803246451
Length 394 pages
Edition 1st Edition
Languages
Tools
Arrow right icon
Author (1):
Arrow left icon
Duc Haba Duc Haba
Author Profile Icon Duc Haba
Duc Haba
Arrow right icon
View More author details
Toc

Table of Contents (17) Chapters Close

Preface 1. Part 1: Data Augmentation
2. Chapter 1: Data Augmentation Made Easy FREE CHAPTER 3. Chapter 2: Biases in Data Augmentation 4. Part 2: Image Augmentation
5. Chapter 3: Image Augmentation for Classification 6. Chapter 4: Image Augmentation for Segmentation 7. Part 3: Text Augmentation
8. Chapter 5: Text Augmentation 9. Chapter 6: Text Augmentation with Machine Learning 10. Part 4: Audio Data Augmentation
11. Chapter 7: Audio Data Augmentation 12. Chapter 8: Audio Data Augmentation with Spectrogram 13. Part 5: Tabular Data Augmentation
14. Chapter 9: Tabular Data Augmentation 15. Index 16. Other Books You May Enjoy

Standard audio augmentation techniques

Similar to image augmentation in Chapter 3, various audio libraries provide many more functions than are necessary for augmentation. Therefore, we will only cover some of the methods available in the chosen audio library.

In image augmentation, the term safe level is defined as not altering or distorting the original image beyond an acceptable level. There is no standard terminology for deforming the original audio signal beyond a permissible point. Thus, the term safe or true will be used interchangeably to denote a limit point for the audio signal.

Fun challenge

Here is a thought experiment: all audio files are represented as numbers in time series format. Thus, can you create a statistically valid augmentation method that does not consider human hearing perception? In other words, use math to manipulate a statistically valid number array, but never listen to the before and after effects. After all, audio augmentation aims to have more...

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