Acoustic Signal Processing in the Age of AI: Methods, Hardware, and Impact

Dear Colleagues,

Advances in machine learning, edge/embedded computing, and cross-domain sensing have reinvented acoustic signal processing. This Special Issue gathers breakthroughs that transform how we analyze, enhance, localize, compress, synthesize, and interpret acoustic and vibroacoustic data—from speech and spatial audio to underwater, aeroacoustic, industrial, and ecological monitoring.

We welcome contributions that demonstrate robustness, latency/throughput gains, energy efficiency, interpretability, privacy, and scalability, beyond accuracy alone. Studies that unite signal-processing theory with modern learning (e.g., transformers, diffusion, neural ODE/SDEs), leverage distributed/federated sensing, or bridge acoustics with robotics, wearable/IoT, ecoacoustics, and smart-city infrastructures are especially encouraged. Real-world deployments, open datasets, and reproducible pipelines are strongly favored.

Article types: Original research, reviews, perspectives. Negative results with rigorous analysis, ablations, and open resources are welcome.

This Special Issue will focus on (but is not limited to) the following topics:

Foundations and methods

• Self-supervised, weakly/few-shot, and domain-adaptation methods for audio/ultrasound/sonar;
• Transformers, diffusion/generative models, neural ODE/SDEs; physics-informed and probabilistic models;
• Time–frequency representations, scattering/wavelet front-ends, robust beamforming and array processing;
• Graph/geometric learning for microphone/sonar networks; uncertainty quantification and calibration;
• Inverse problems, sparse/low-rank optimization, compressive and model-based deep learning.

Systems, hardware and edge

• TinyML, quantization/pruning/distillation; neuromorphic and spiking approaches;
• FPGA/ASIC/DSP implementations; scheduling for energy-aware, low-latency inference;
• Federated/split learning, privacy-preserving analytics; Internet of Acoustic Things (IoAT).

Applications

• Speech enhancement/separation/bandwidth extension; robust ASR; spatial and immersive audio (AR/VR);
• Bio/ecoacoustics across marine, terrestrial, and urban soundscapes; autonomous UUV/USV/UAV sensing;
• Underwater acoustics (passive/active), marine mammal/fish detection, geo/acoustic mapping;
• Aero/industrial/structural health monitoring; acoustic emission and vibroacoustics; HCI and assistive hearing.

Trustworthy and responsible acoustic AI

• Explainability (e.g., SHAP/LIME/attribution), fairness, safety under distribution shift, adversarial robustness;
• Data governance, annotation quality, simulation-to-real transfer, ethical considerations.

Data, benchmarking and reproducibility

• Open datasets, synthetic corpora, task protocols; evaluation beyond accuracy (latency, memory, joules/inference);
• Reproducible pipelines, MLOps for acoustic ML, standardized reporting and ablation practices.

Dr. Pai-Chen Guan
Prof. Dr. Jui-Hsiang Kao
Dr. Shashidhar Siddagangaiah
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 100 words) can be sent to the Editorial Office for announcement on this website.

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-blind 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.

• acoustic AI
• acoustic signal processing
• audio machine learning
• self-supervised learning
• transformers
• diffusion models
• beamforming
• source separation
• speech enhancement
• graph neural networks
• TinyML
• edge AI
• Internet of Acoustic Things
• underwater acoustics
• ecoacoustics
• spatial/immersive audio
• robustness
• explainability
• neuromorphic computing