Skip to content

Geometry-adaptive Ambisonic encoding for sparse microphone arrays of variable topology using physics-informed diffusion

Aug 2026 · arXiv.org · Vol abs/2608.16240 · 1 citation · 37 references
Computer Science Engineering

TL;DR

DiffM2A is presented, a geometry-adaptive conditional diffusion framework for robust Ambisonic encoding from sparse MAs with variable topologies that outperforms conventional and neural baseline methods on signal fidelity, spectral accuracy, spatial coherence, and binaural cue preservation.

Abstract

Ambisonics delivers compact scene based spatial audio representation, yet higher order Ambisonic encoding poses difficulties for wearables and embedded hardware. Their microphone arrays are often sparse, irregular, and constrained by device specific boundary conditions. These factors make the spherical-harmonic (SH) domain encoding ill conditioned: inverse filtering amplifies noise, while deterministic neural encoders may overfit to array-specific responses or smooth ambiguous higher-order components. This paper presents DiffM2A, a geometry-adaptive conditional diffusion framework for robust Ambisonic encoding from sparse MAs with variable topologies. Its Geometry-Adaptive Spherical Harmonic Projection (GASHP) front-end constructs boundary-aware SH steering functions and applies an energy-normalized modal projection, mapping array-dependent observations to a common modal representation without explicit pseudo-inverse computation. A dual-branch Elucidated Diffusion Model then estimates complex Ambisonic coefficients, conditioned on both the raw microphone spectra and GASHP features. Sound intensity and rotational equivariance losses further enhance inter-channel phase consistency and structured behavior across SH subspaces. Evaluations on both first- and second-order Ambisonic encoding tasks, using simulated room-acoustics and real-world LOCATA recordings, demonstrate that DiffM2A outperforms conventional and neural baseline methods on signal fidelity, spectral accuracy, spatial coherence, and binaural cue preservation. Additional experiments show that these gains are largely retained across unseen five-microphone layouts and under mismatched open-array and rigid-sphere boundary models.

View source

Similar papers

Preprint Oct 2026

Array-Agnostic High-Order Ambisonics Encoding via Diffusion Posterior Sampling

Ambisonics provides a device-independent representation of three-dimensional sound fields; however, extracting these coefficients from practical microphone array recordings remains strongly dependent on array geometry, acoustic transfer functions, and spatial sampling resolution. This pipeline is typically divided into...

Amit Milstein, Nir Shlezinger, B. Rafaely · 0 citations
Preprint Sep 2026

Signal-Independent and Signal-Dependent Neural Ambisonic Matrix Encoding for Arbitrary Arrays with Variable Microphone Counts

Recent neural Ambisonic encoders accommodate diverse array geometries, yet many existing neural encoders require a fixed microphone count because the number of microphone channels is embedded in the network architecture. This requirement limits deployment across devices with different microphone configurations and adap...

Shi-Chao Hu, Zhi-Heng Jin, Chun-Yang Xu et al. · 0 citations
Open access Oct 2026

A 4-D spatio-temporal misfit function for waveform inversion of Distributed Acoustic Sensing data

Standard Full-Waveform Inversion (FWI) objective functions often struggle to resolve deep structures when data are dominated by localised near-surface scattering. To address this, we develop a 4-D time-frequency-space-wavenumber (TFXK) objective function that explicitly enforces spatial coherence, leveraging the dens...

E. Shimony, A. Lellouch · 0 citations
Open access 2026

A Hybrid Beamforming-Post-Filter Framework for Ambisonics Enhancement

Ambisonics recordings encoded from multi-microphone signals enable effective capture of three-dimensional soundfield information, thus have been widely adopted in immersive media, virtual reality, and augmented reality applications. However, in real-world scenarios, the directly acquired Ambisonics signals typically co...

Liang Tao, Mao-Shen Jia, Yong-Gang Hu · 0 citations
Open access Oct 2026

Rethinking Channel Estimation for Fluid Antenna Systems: A Physics-Informed Operator Learning

Fluid antenna systems (FAS) leverage continuous reconfigurable apertures to transcend the spatial limits of fixed-position arrays, yet acquiring full-aperture channel state information (CSI) from sparse observations remains a formidable bottleneck. In this paper, we establish a continuous-field formulation for FAS by r...

Zi-Xin Wang, Li-Zhong Zheng, Yuan-Ming Shi et al. · 1 citation
Preprint Sep 2026

GAMF: Learned and Analytical Array Transfer Function Matching for Array-Generic Direction-of-Arrival Estimation

Simulations across array configurations show that both learned and hybrid configurations outperform a representative positional-encoding-based neural baseline, remain competitive with analytical ATF matching in clean, low-reverberation scenes, and substantially improve upon it under stronger noise or reverberation.

Zhi-Heng Jin, Shi-Chao Hu, Chun-Yang Xu et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.