Neural Audio Plugin Competition
Supported by
Deep learning has provided numerous breakthroughs within the creative industries. Harnessing this creative potential within audio plug-ins provides an exciting intersection between academia and the music tech industry.
Neutone and The Audio Programmer have partnered to create the Neural Audio Plug-in Competition (NAP). This initiative will allow the creators of neural audio effects to demonstrate their potential and help empower artists with the latest advances in this groundbreaking technology.
We invite your participation and support for this non-profit event! Join our Discord here for updates on the competition! Here are the basics…
Who is this Competition For?
Anyone who can create a neural audio plug-in can enter. We also allow teams of up to 4.
How Can I Create a Neural Audio Plug-in?
There have been some incredible examples of Neural Audio plugins that have been released in recent years including DDSP-VST, Mawf, Steinberg’s Backbone using Drum GAN, Neutone, as well as various others!
You may be wondering how you can create your own Neural Audio Plugin! Below are several powerful tools, SDK’s and resources that you can use to create your own neural audio plug-in.
How do I Submit a Plug-in?
Once the competition opens, you can submit by supplying a repository link to your entry. We are accepting open source submissions only, as one of the criteria for the competition will be the quality of the source code. Make a submission here.
When will the Competition be Held?
The competition will run from 12:01 am GMT on 14 February 2023 (the "Opening Date") to 11:59 pm GMT on 24 March 2023 (the "Closing Date") inclusive.
In the preceding 4 weeks leading up to the event, we will be holding live-streamed workshops, demos and discussions on neural audio.
What are the Terms and Conditions?
- Teams of up to 4 can participate
- Only open source submissions will be admitted for entry
- Submissions need to be in VST3 or AU audio plug-in format
- Submissions must use a deep learning or machine learning algorithm
- Prizes are to be split amongst the team. Teams will decide for themselves how the prizes will be distributed. Any bank wire fees will be deducted from the cash prize.
Wait…There are Prizes for Winners?
Yes! The competition will be judged based on the following perceptual and technical criteria:
- Novelty - Innovation in the machine learning algorithm or application
- Usefulness - How much impact the plugin has on improving creative workflow/ideas
- Character - A measure of uniqueness, quirkiness and enjoyment through UI/UX
- Versatility - The variety of different use-cases the plugin caters for. And/or the variety of content the plugin can produce.
- Performance - Technical performance, latency, responsiveness and stability of the plugin
Prizes
1st Place
Courtesy of
- $2000 - courtesy of TikTok
- JUCE indie or Max & RNBO license for each member (up to 4)
- NAMM Ticket and Presenting opportunity (GPU Audio)
2nd Place
Courtesy of
- $1000 - courtesy of TAP & Qosmo
- JUCE indie or Max & RNBO license for each member (up to 4)
3rd Place
Courtesy of
- $500 - courtesy of GPU Audio
Recommended SDKs & Frameworks
Here are some popular tools to help you create your Neural Audio Plug-in! Please feel free to use another SDK or Framework for your submission.
CMajor
CMajor is a programming language for writing fast, portable audio software. You've heard of C, C++, C#, objective-C... well, Cmajor is a C-family language designed specifically for writing DSP signal processing code.
iPlug2
Originally created by Cockos, iPlug is an awesome audio plug-in framework. iPlug2 has support for GPU accelerated vector graphics, resizable high DPI UIs, iOS, WebViews, AUv3 and compiling to WebAudio/WASM, amongst many other things. iPlug2's syntax is super simple, for example, creating a parameter or a control in the UI is only a single line of C++ code.
JUCE
JUCE is a widely used framework for audio application and plug-in development. It is an open source C++ codebase that can be used to create standalone software on Windows, macOS, Linux, iOS and Android, as well VST, VST3, AU, AUv3, AAX and LV2 plug-ins.
MATLAB
MATLAB is a programming and numeric computing platform used by millions of engineers and scientists to analyze data, develop algorithms, and create models. Get a provisional license for the competition here. The Audio Toolbox enables creating VST plugins from MATLAB code.
Neutone
AI audio plugin & community bridging the gap between AI research and creativity. For creators, Neutone makes AI technologies accessible for all to experiment with. You’ll find transformative AI audio instruments that will spark endless creative possibilities. For researchers, Neutone is a go-to platform for you to share real-time AI audio processing models with potential users in the audio production community.
RTNeural
RTNeural is a lightweight neural network inferencing engine written in C++. This library was designed with the intention of being used in real-time systems, specifically real-time audio processing.
Submission Requirements
How to enter:
- 1.Visit the Recommended SDKs & Frameworks for tools and resources for building your Neural Audio Plugin.
- 2.Develop your Neural Audio VST3 or AU plugin using an available SDK or by deploying your Neural Audio model through Neutone or other model deployment framework within a DAW environment.
- 3.Record a video demonstrating your design process and finished product. Please show your VST3/AU being used as part of a DAW project.
- 4.Commit your code to an open-source Git repository(MIT or Apache License) and submit a URL to the repo. Provide a Read Me with some basic documentation and build instructions.
- 5.Submit all your material through the sign up link provided.
- 6.Please enter independently or as a team of up to four members. There is no entry fee to participate.
Organizing Team
Christian Steinmetz
Queen Mary University of London
Joshua Hodge
The Audio Programmer
Andrew Fyfe
Qosmo
Have a question?
Please join the Audio Programmer Discord and ask us in "Neural Audio Plug-in Competition"