Forum Acusticum 2026

Session

A09.00 Machine learning and artificial intelligence in acoustics

A09.00

11 Sept 2026, 08:40

Conveners

A09.00 Machine learning and artificial intelligence in acoustics: S053

• Martin Hagmüller (Signal Processing and Speech Communication Laboratory)
• Alois Sontacchi (University of Music and Performing Arts)
• Mirco Pezzoli
• Franz Pernkopf (Signal Processing and Speech Communication Laboratory)

A09.00 Machine learning and artificial intelligence in acoustics: P509

• Franz Pernkopf (Signal Processing and Speech Communication Laboratory)
• Mirco Pezzoli
• Alois Sontacchi (University of Music and Performing Arts)
• Martin Hagmüller (Signal Processing and Speech Communication Laboratory)

A09.00 Machine learning and artificial intelligence in acoustics: S386

• Mirco Pezzoli
• Alois Sontacchi (University of Music and Performing Arts)
• Martin Hagmüller (Signal Processing and Speech Communication Laboratory)
• Franz Pernkopf (Signal Processing and Speech Communication Laboratory)

A09.00 Machine learning and artificial intelligence in acoustics: S388

• Mirco Pezzoli
• Martin Hagmüller (Signal Processing and Speech Communication Laboratory)
• Alois Sontacchi (University of Music and Performing Arts)
• Franz Pernkopf (Signal Processing and Speech Communication Laboratory)

4015. A Deep Neural Network for Predicting the Sound Field Radiated by Orthogonally Stiffened Plates (FA2026/220)

Manuele Montrasio (Politecnico di Milano)

11/09/2026, 08:40

A09 Machine learning and artificial intelligence in acoustics

In recent years, the adoption of surrogate modeling in vibroacoustics applications is being explored. Artificial Neural Networks (ANNs) are proving their effectiveness in predicting structural vibration, and more recently their capabilities in sound radiation prediction is being investigating as well, paving the way for new research in the field. This paper explores the use of a Deep Neural...

4016. HERMES: A Physics-informed Hierarchical Mesh Transformer for Real-time Room Eigenfrequency Prediction (FA2026/202)

Bence Bakos (Eötvös Loránd University)

11/09/2026, 09:00

A09 Machine learning and artificial intelligence in acoustics

The accurate prediction of low-frequency room acoustics relies on modal frequencies, traditionally computed using expensive numerical methods such as the Finite Element Method (FEM). This paper presents HERMES (Hierarchical Estimator for Room Modal Eigenvalue Synthesis), a physics-informed hierarchical mesh transformer that predicts the first 200 modal frequencies directly from 3D surface...

4019. murenn: Multi-Resolution Neural Networks in PyTorch (FA2026/394)

Xiran Zhang (Nantes University)

11/09/2026, 09:00

A09 Machine learning and artificial intelligence in acoustics

Multi-Resolution Neural Networks (MuReNN) are a new generation of models for deep learning for speech and audio processing. Compared to convolutional networks (convnets), they are more resource-efficient, less sensitive to initialization, and may generalize better across recording conditions. The key idea behind MuReNN is to learn a filterbank in which each filter is factorized between a...

4017. Two-stage sound field learning through holomorphic neural networks and Vekua operators (FA2026/437)

Matteo Calafà (DTU Electro)

11/09/2026, 09:20

A09 Machine learning and artificial intelligence in acoustics

Holomorphic neural networks have recently been proposed as two-dimensional universal function approximators constrained to produce holomorphic outputs. In this work, we combine holomorphic neural networks with Vekua operators toconstruct a machine learning architecture that inherently generates solutions tothe Helmholtz equation. This approach is particularly advantageous for two-dimensional...

4018. A neural eigenvalue solver using plane wave representations (FA2026/440)

Matteo Calafà (DTU Electro)

11/09/2026, 09:40

A09 Machine learning and artificial intelligence in acoustics

Estimating acoustic eigenfrequencies is essential in the design of rooms and buildings, particularly at low frequencies and/or small rooms. While analytical solutions exist for simple geometries, more general shapes and materials require numerical algorithms. In recent years, machine learning approaches for eigenvalue problems have been proposed, but their accuracy and robustness remain...

4020. Dynamic Sound Field Regression via Physics-informed Temporal- Frequency Residual Learning (FA2026/656)

Yuanxin Xia (DTU Electro)

11/09/2026, 10:20

A09 Machine learning and artificial intelligence in acoustics

Dynamic sound sources, such as footsteps or other moving sources, are common in real life but difficult to model because their characteristics and trajectory are often unknown. This paper presents a machine learning algorithm for sound field construction based on a temporal-frequency residual neural network. The algorithm is designed to capture how sound evolves across both time and frequency,...

4021. Real-Time Broadband Sound Field Regression via Massively Parallel Frequency Stacking Neural Networks (FA2026/716)

Yuanxin Xia (DTU Electro)

11/09/2026, 10:40

A09 Machine learning and artificial intelligence in acoustics

Neural implicit fields are a powerful framework for sound field regression, but their application to broadband problems has been challenged by the prohibitively expensive computational cost of training neural networks. To break this barrier, this paper proposes a massively parallel strategy via frequency stacking. This architecture factorizes the broadband problem into independent,...

4022. Topological Features for Environmental Sound Classification (FA2026/122)

Jiwon Seo (Fraunhofer IDMT)

11/09/2026, 11:00

A09 Machine learning and artificial intelligence in acoustics

In this work, we investigate whether topological features derived from persistent homology of Takens embeddings can complement conventional acoustic features for environmental sound classification on the ESC-50 dataset. A compact statistical summary of persistence diagrams is extracted from short overlapping segments, aggregated at the recording level, and combined with conventional acoustic...

4023. AI-Driven Detection and Acoustic Modeling of Anthropogenic Underwater Noise Across European Marine Basins (FA2026/258)

Rosa Ma Alsina-Pagès (La Salle, Universitat Ramón Llull)

11/09/2026, 11:20

A09 Machine learning and artificial intelligence in acoustics

Anthropogenic underwater noise is becoming an increasing concern for marine ecosystems, especially in coastal and semi-enclosed areas exposed to intense commercial, industrial, and recreational vessel traffic. Within the DeuteroNoise project, we present an AI-driven framework to detect, characterize, and model vessel-generated underwater noise across five European marine basins: the North...

4024. Application of Siamese networks to Anomaly Detection in Multibeam Sonograms For CCS Monitoring (FA2026/787)

Jonathan Farr (University of Southatmpon)

11/09/2026, 11:40

A09 Machine learning and artificial intelligence in acoustics

Effective anomaly detection in active sonar data is critical for the long-term monitoring of offshore Carbon Capture and Storage (CCS) sites, where early identification of leaks is essential for environmental safety and regulatory compliance. Multibeam sonar systems offer high-resolution imaging of the water column but produce large amounts of data, rendering conventional reconstruction-based...

4025. Acoustic classification of packaging waste based on ultrasonic and audible sound wave emissions in reverberant enclosures (FA2026/471)

Marc Arnela (La Salle, Universitat Ramón Llull)

11/09/2026, 14:40

A09 Machine learning and artificial intelligence in acoustics

Reverse vending machines encourage recycling, but require automatic container classification. This is typically achieved by reading barcodes or using computer vision, which can fail due to unreadable codes or poor lighting. Recently, an active acoustic approach using sound waves has been proposed as an alternative, offering low computational cost and minimal maintenance. However, its...

4026. Interpretable Machine Learning for Assessing Environmental and Mounting Effects on Aeronautical Glass Wool Absorption (FA2026/277)

Alfonso Caiazzo (University of Naples Federico II)

11/09/2026, 15:00

A09 Machine learning and artificial intelligence in acoustics

The in-service acoustic performance of fibrous absorbers in aircraft fuselage linings is affected by installation variability and environmental exposure, yet their relative contributions remain poorly quantified. This work reframes the problem as a feature detectability task, where the sound absorption coefficient is used to infer the physical parameters governing its variability. Two...

4027. SPSC-HCM-16C: A Multi-Channel Lung Sound Benchmark (FA2026/179)

Benedikt Mayrhofer (Signal Processing and Speech Communication Laboratory)

11/09/2026, 15:20

A09 Machine learning and artificial intelligence in acoustics

This paper presents SPSC-HCM-16C, a benchmarkdataset1for patient-level respiratory disease classification from synchronized 16-channel lung soundrecordings collected in a real clinical environment.The dataset contains recordings from 183 subjects,including healthy controls and four respiratory diseasegroups, and supports three diagnostic settings: 2-class,3-class, and 5-class classification....