Exploring Contextual Timbre Representation
Date:
Wed, 07/24/2024 - 3:00pm - 4:00pm
Location:
CCRMA Classroom
Event Type:
Guest Lecture 
Given that most music consumption involves multi-instrument audio, understanding complex timbres in context is vital for real-world music analysis and Music Information Retrieval (MIR) tasks. Modeling instrument timbre is crucial in MIR, yet remains challenging in musical mixtures where definitions of timbre are less clear. Additionally, the advent of generative and multimodal music models necessitates precise timbre representation for enhanced control in music generation and editing. This talk explores the challenges in modeling predominant instrumentation, revealing the fragility of existing discriminative solutions. We present a study on using diffusion models to generate predominant instrumentation, aimed at exploring the potential of generative models to deepen our understanding of instrument representation for scenarios with limited real-world data.
NOTE: this event is part of the DL4MIR workshop series (ccrma-mir.github.io); guest speaker talks are open to the broader CCRMA community.
FREE
Open to the Public
