Thu, Oct 1 · 6:00 PM CEST
On the 1st of October we will host the next iteration of our monthly Audio Developer Meetup. Fabián Esqueda from KORG will dig into audio circuit modeling and Tim Bencker will be presenting about what audio developers can learn from other domains. Anyone interested in Audio development is welcome!
Doors open at 18:00, talks start at 18:15. No admission after 18:30.
Please note: This month, we may have less capacity due to ongoing renovations; we will only know very last-minute whether we will have access to the usual room. So please check back regularly if you are on the waiting list! Attendance is strictly limited to people who have RSVP'd.
Talk Details:
The Grayscale of Audio Circuit Modeling: White-Box, Black-Box, and the Space Between (Fabián Esqueda)
Circuit modeling, or virtual analog, is a well-established topic in music technology. Over its 30-year history, it has produced a substantial body of research and products, bringing the sounds of legendary vintage instruments to the personal computers of musicians and hobbyists. The field is often framed as a divide between white-box approaches, which derive models from a circuit's physical equations, and black-box approaches, which learn behavior from measured data. In practice, most modeling work lives in the gray area between them. Virtual analog continues to evolve and, perhaps unsurprisingly, machine learning has injected new life into it. This talk gives an overview of the field, then dives into personal contributions and experiences, contextualized with references to recent KORG releases.
Audio's Problems Aren't Unique - and That's Good News (Tim Bencker)
Building audio software is hard, but many of the hard parts aren't audio. Complexity grows, requirements change, and unresolved design decisions come back to haunt you. Game and web development face similar problems and have developed practical ways to handle them. Drawing on work between design and engineering, this talk looks at what audio can borrow: how to structure products and systems, how to make decisions, and what to automate. AI makes code faster to write and harder to keep track of. The same techniques that help teams move fast without losing control turn out to work just as well for AI agents.
About the Speakers:
Fabián Esqueda is a Research Engineer at KORG, where he works on signal processing and machine learning algorithms for musical instruments. His research interests include generative models, sound synthesis, and virtual analog modeling. He holds an MSc in Acoustics and Music Technology from the University of Edinburgh, UK, and a DSc from Aalto University, Finland, where he specialized in antialiasing techniques for nonlinear signal processing. Prior to joining KORG, he worked as a DSP and ML Research Engineer at Native Instruments. In his spare time, he can usually be found at the Berlin Zoo with his 2-year-old daughter.
Tim Bencker is a designer and developer. Since 2021 he has built interactive AR and real-time applications at Sensape for brands such as Warner Bros, first as technical artist and then as project manager. He is part of the team behind Scyclone, the neural timbre transfer plugin that won the 2023 Neural Audio Plugin Competition, and builds Oudio.io, a platform for selling and licensing audio software. He is also speaking at ADC 2026 in Bristol.
