Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Learn how to create a custom audio dataset using PyTorch and torchaudio in this comprehensive tutorial video. Explore the process of loading audio files from the UrbanSound8K dataset while gaining insights into basic I/O functions in torchaudio. Dive into the implementation of a custom UrbanSoundDataset class, including the constructor, __len__, and __getitem__ methods. Understand how to work with PyTorch's Dataset and DataLoader classes, navigate the UrbanSound8K dataset structure, and utilize torchaudio's backend for efficient audio file handling. Follow along as the instructor demonstrates the entire process, from initial setup to running the completed UrbanSoundDataset class, providing valuable hands-on experience in audio data preprocessing for machine learning applications.
Syllabus
Intro
Urban Sound 8K Dataset webpage
PyTorch Dataset and DataLoader classes
UrbanSoundDataset class
Implementing the constructor - part 1
Content of Urban Sound 8K
Implementing the constructor - part 2
Implementing __len__
Implementing __getitem__ - part 1
Torchaudio backend I/O
Implementing __getitem__ - part 2
Getting audio sample path
Getting label
UrbanSoundDataset class recap
Running UrbanSoundDataset
Coming up next
Taught by
Valerio Velardo - The Sound of AI