Data-Driven and Physics-Inspired Methods in Audio and Music Signal Processing

Dear Colleagues,

In recent years, machine learning and deep neural networks have reshaped audio and music signal processing, enabling unprecedented advances in sound synthesis and processing, music analysis, and audio signal enhancement. These data-driven methods have unlocked new possibilities for solving extremely challenging tasks, leading to scenarios and applications that were unthinkable just a few years ago. However, despite this remarkable progress, purely data-driven approaches often come with limitations: they can lack interpretability, require large datasets and computational resources, and may struggle to reliably generalize beyond training conditions.

Meanwhile, physics-informed and traditional digital signal processing (DSP) methods have continued to play an important role in audio signal processing and are once again becoming the focus of growing interest. These approaches offer clear structure and interpretability and can benefit significantly from being optimized or combined with data-driven techniques. Indeed, some of the most interesting advances in the field are now arising from strategies that integrate learning-based models with principled DSP and physical modeling. Differentiable DSP frameworks and neural–physical hybrid audio processing pipelines have already demonstrated that this synergy can lead to methodologies able to unify physical accuracy with the power of neural models, paving the way for more controllable, explainable, expressive, and robust audio/music technologies.

This Special Issue, “Data-Driven and Physics-Inspired Methods in Audio and Music Signal Processing”, aims to highlight advances that bridge the gap between classical DSP and modern machine learning approaches. We invite contributions that explore differentiable signal processing, data-efficient learning, and physics-inspired neural architectures and rigorously analyze the trade-offs between learning models, traditional systems, and classical baselines. We particularly welcome work that explicitly integrates physical structures or DSP priors into learning systems, as well as work that addresses both theoretical foundations and practical applications (e.g., deployment scenarios, rigorous latency evaluation, and hardware-aware machine learning approaches) in music technology, audio engineering, and sound computing.

Original research articles and reviews are welcome. Substantially extended versions of relevant conference papers may also be considered. Topics of interest include, but are not limited to:

• Virtual analog modeling;
• Neural modeling of physical systems;
• Differentiable digital signal processing;
• Physics-inspired machine learning for audio and music applications;
• Real-time implementations and embedded systems for audio and music processing;
• Physics-informed soundfield reconstruction;
• Spatial audio and immersive systems grounded in physical modeling;
• Explainable and interpretable AI for sound analysis, synthesis, and processing;
• Self-supervised and data-efficient learning systems with structural constraints;
• Generative models with physical priors or an interpretable structure.

We look forward to receiving your contributions to this exciting and rapidly evolving field.

Dr. Riccardo Giampiccolo
Dr. Alessandro Ilic Mezza
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the Special Issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-anonymized peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.