C4DM Seminar: Sungho Lee
QMUL, School of Electronic Engineering and Computer Science
Centre for Digital Music Seminar Series
Seminar by:
Sungho Lee
Date/time: 16th of November, 11:00-12:00
Location: G2, Mile End Campus
Open to students, staff, alumni, public; all welcome. Admission is FREE, no pre-booking required.
Title: Deep learning meets audio processing graphs
Abstract: Abstract: Musicians and audio engineers sculpt and transform their sounds by connecting multiple processors, forming an audio processing graph. However, most deep-learning methods overlook this real-world practice and assume fixed graph settings. In this talk, we will share our attempts to fill this gap. We will first introduce our recent proof-of-concept, which trains a graph transformer that estimates the original graph (which processors exist, how they connect to others, and what parameter values each processor has) from reference audio. Then, we will outline our recent ongoing extensions, which include efficient computation of arbitrary audio processing graphs on GPU and differentiable search (reverse engineering) of music mixing systems.
Read more: https://arxiv.org/abs/2303.08610