Deep Learning In Audio for Absolute Beginners: From No Experience & No Datasets to a Deployed Model
This video has been re-uploaded due to an error in the original
https://audio.dev/ -- @audiodevcon
Jumpstart Guide to Deep Learning In Audio for Absolute Beginners: From No Experience and No Datasets to a Deployed Model - Jan Wilczek - ADC22
Deep learning is becoming more and more significant in all areas of science including audio processing. Yet a lot of people have a hard time understanding it or are too scared to start learning it altogether. Is deep learning really so difficult that you need a PhD to use it? Does it truly require huge datasets and gigantic computational clusters? Is it possible to deploy neural networks in real-time audio plugins? In this talk, I will show you how you can learn deep learning for Virtual Analog modeling of audio effects fast, for free, without a PhD, without any special equipment or loads of data, and deploy your deep learning model in an audio plugin.
What you will learn:
what are the 4 biggest myths concerning deep learning?
how to learn deep learning for audio fast in 4 simple steps for free
where to find and how to synthesize a dataset to model your analog device of choice
how to train your first deep learning model for audio using the basics of PyTorch and without a computational cluster
how to deploy your model in a real-time audio plugin
The presentation will feature a live demo of setting up a deep learning pipeline and training a neural network for Virtual Analog modeling of a distortion effect.
Slides: link will be updated when available.
_
Jan Wilczek
I am an Audio Developer of Music Maker JAM at Loudly GmbH in Berlin; an app to make loop-based music that runs on Android, iOS, and Windows.
Additionally, I have created https://TheWolfSound.com to help students and software engineers learn audio programming for getting game audio, audio plugin, or mobile audio developer jobs. I am also a researcher in the area of Virtual Analog modeling using deep learning; my latest work was presented at the Digital Audio FX 2022 conference in Vienna. My professional goal is to lower the entry barrier for everyone interested in audio. I pursue it by publishing high-quality articles, videos, and podcasts or... speaking at conferences.
_
Streamed & Edited by Digital Medium Ltd - online.digital-medium.co.uk
_
Organized and produced by JUCE: https://juce.com/
_
Special thanks to the ADC22 Team:
Lina Berzinskas
Sophie Carus
Derek Heimlich
Andrew Kirk
Bobby Lombardi
Tom Poole
Ralph Richbourg
Jim Roper
Jonathan Roper
#audiodevcon #audiodev #deeplearning