Generating Sound Digits with a Variational Autoencoder
Offered By: Valerio Velardo - The Sound of AI via YouTube
Course Description
Overview
Learn how to use a trained Variational Autoencoder to generate spectrograms of sound digits and convert them into audio in this 35-minute tutorial video. Explore the implementation of a SoundGenerator, create a generation script, troubleshoot bugs, and listen to the resulting generated sound digits. Access accompanying code and slides for hands-on practice. Part of the "Generating sound with neural nets" series, this video provides practical insights into advanced audio generation techniques using machine learning.
Syllabus
Intro
Sound generation with VAEs
Implementing SoundGenerator
Generate script
Bug hunting + fixing
Listening to generated sound digits
Outro
Taught by
Valerio Velardo - The Sound of AI
Related Courses
Basics of Noise and Its MeasurementsIndian Institute of Technology Kanpur via Swayam Audio Classification with TensorFlow
Coursera Project Network via Coursera Fourier Analysis
Steve Brunton via YouTube Audio Analytics with Azure Automated ML for Music Genre Classification
Microsoft via YouTube Tagging Audio Using Deep Learning
Microsoft via YouTube