Uploaded April 2021 | Updated September 2026, 2 weeks ago

---

In the last installment of the tutorial series “Generating sound with neural nets”, you can learn how to use a previously trained Variational Autoencoder to generate spectrograms of sound digits
convert the generated spectrograms into audio.

Code:
github.com/musikalkemist/generating-sound-with-neural-networks/tree/main/14%20Sound%20generation%20with%20VAE/code

Slides:
github.com/musikalkemist/generating-sound-with-neural-networks/tree/main/14%20Sound%20generation%20with%20VAE/slides

===============================

Interested in hiring me as a consultant/freelancer?
valeriovelardo.com

Join The Sound Of AI Slack community:
valeriovelardo.com/the-sound-of-ai-community

Follow Valerio on Facebook:
facebook.com/TheSoundOfAI​​​

Connect with Valerio on Linkedin:
linkedin.com/in/valeriovelardo

Follow Valerio on Twitter:
twitter.com/musikalkemist​​​​​​​

===============================

Content:
0:00 Intro
1:01 Sound generation with VAEs
4:25 Implementing SoundGenerator
16:23 Generate script
27:14 Bug hunting + fixing
32:33 Listening to generated sound digits
34:45 Outro