About this Research Topic
In recent years, physical modeling has seen a growing body of scientific works in music and audio applications. This approach aims to synthesise or manipulate sound starting from the mathematical description of musical instruments and devices such as audio circuits.
Physical modelling offers several advantages including flexibility, the ability to replicate natural transients and articulations by setting a few parameters, and establishing a causal relationship between inputs and sound output. These attributes have helped consolidate the use of physics-based sound synthesis and processing both scientifically and artistically.
Model-based synthesis distinguishes itself from signal-based methods, in that it enhances our understanding of the mechanisms behind sound production, providing interpretability and insight.
Physical modeling synthesis has expanded considerably from its original applications, thanks to continuous refinement of the underlying phenomena it seeks to emulate. Common challenges encountered in current research trends include the simulation of multi-physics systems; nonlinear phenomena; large-scale problems; time-varying parameters.
To address these complexities, researchers have developed advanced modeling, methods (deterministic or sometimes stochastic) and numerical schemes with guaranteed properties regarding accuracy, convergence, passivity, and aliasing suppression. These numerical techniques not only enhance the sonic realism but also improve computational efficiency, making them more accessible for real-time performance and production environments.
Machine learning and deep learning applications have considerably expanded the landscape of the available techniques in parameter identification and optimisation.
The aim of this special issue is to collect a series of papers addressing the challenges at the forefront of physics-based audio synthesis and processing. As such, this issue encourages submissions dealing with the most recent developments in the field, with a particular focus on complex systems such as multi-physics, nonlinear or time-variant systems, both at the modeling and the simulation level.
Authors are invited to submit contributions incorporating one or more of the following topics:
-multi-physics modelling
-time-variant parameters
-power-balanced / energy-stable schemes
-energy quadratisation / energy factorisation
-linearly implicit, semi-implicit and explicit schemes
-aliasing suppression techniques
-parameter identification / differentiable physical models
-physics-informed neural networks
-order reduction techniques (eigen decomposition, statistical physics)
-vectorisation, parallelisation, and hardware acceleration
-automatic generation of model equations and/or simulation codes from netlists of multi-physics components and connection graphs
Physical modelling offers several advantages including flexibility, the ability to replicate natural transients and articulations by setting a few parameters, and establishing a causal relationship between inputs and sound output. These attributes have helped consolidate the use of physics-based sound synthesis and processing both scientifically and artistically.
Model-based synthesis distinguishes itself from signal-based methods, in that it enhances our understanding of the mechanisms behind sound production, providing interpretability and insight.
Physical modeling synthesis has expanded considerably from its original applications, thanks to continuous refinement of the underlying phenomena it seeks to emulate. Common challenges encountered in current research trends include the simulation of multi-physics systems; nonlinear phenomena; large-scale problems; time-varying parameters.
To address these complexities, researchers have developed advanced modeling, methods (deterministic or sometimes stochastic) and numerical schemes with guaranteed properties regarding accuracy, convergence, passivity, and aliasing suppression. These numerical techniques not only enhance the sonic realism but also improve computational efficiency, making them more accessible for real-time performance and production environments.
Machine learning and deep learning applications have considerably expanded the landscape of the available techniques in parameter identification and optimisation.
The aim of this special issue is to collect a series of papers addressing the challenges at the forefront of physics-based audio synthesis and processing. As such, this issue encourages submissions dealing with the most recent developments in the field, with a particular focus on complex systems such as multi-physics, nonlinear or time-variant systems, both at the modeling and the simulation level.
Authors are invited to submit contributions incorporating one or more of the following topics:
-multi-physics modelling
-time-variant parameters
-power-balanced / energy-stable schemes
-energy quadratisation / energy factorisation
-linearly implicit, semi-implicit and explicit schemes
-aliasing suppression techniques
-parameter identification / differentiable physical models
-physics-informed neural networks
-order reduction techniques (eigen decomposition, statistical physics)
-vectorisation, parallelisation, and hardware acceleration
-automatic generation of model equations and/or simulation codes from netlists of multi-physics components and connection graphs
Keywords: multi-physics modelling, time-variant parameters, energy quadratisation, energy factorisation, aliasing suppression techniques, parameter identification, differentiable physical models, neural networks, vectorisation, parallelisation
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
