First-order Ambisonics (FOA) is a standard spatial audio format based on spherical harmonic decomposition. Its zeroth- and first-order components capture the sound pressure and particle velocity, respectively. Recently, physics-informed neural networks have been applied to the spatial interpolation of FOA signals, regularizing the network outputs based on soft penalty terms derived from physical principles, e.g., the linearized momentum equation. In this paper, we reformulate the task so that the predicted FOA signal automatically satisfies the linearized momentum equation. Our network approximates a scalar function called velocity potential, rather than the FOA signal itself. Then, the FOA signal can be readily recovered through the partial derivatives of the velocity potential with respect to the network inputs (i.e., time and microphone position) according to physics of sound propagation. By deriving the four channels of FOA from the single-channel velocity potential, the reconstructed signal follows the physical principle at any time and position by construction. Experimental results on room impulse response reconstruction confirm the effectiveness of the proposed framework.
Figures & tables
Figure 1: Overview of the proposed VPNF, where bold solid arrows indicate automatic differentiation. We predict the zeroth- and first-order components of Ambisonics by computing the partial derivatives of the potential with respect to time and position, respectively.
Figure 2: Overhead view of the room geometry for simulation.
Figure 3: NMSE [dB] averaged over ten rooms with different numbers of measurements.
Figure 4: Average Pearson’s correlation coefficients for the W - and (X,Y,Z) -channels.
NMSE [dB] ( ↓ )
Correlation ( ↑ )
W -ch
(X,Y,Z) -ch
W -ch
(X,Y,Z) -ch
100
200
100
200
100
200
100
200
DANF [ 39 ]
−0.38
−0.37
−0.40
−0.43
0.31
0.38
0.31
0.36
PI-DANF [ 30 ]
−2.03
−3.87
−1.52
−2.99
0.62
0.77
0.53
0.69
VPNF
−3.97
−5.04
−3.26
−4.15
0.75
0.81
0.69
0.75
VPNF-Wave
−3.83
−4.74
−3.29
−4.09
0.74
0.79
0.69
0.74
Table 1: Reconstruction results from {100,200} measurements on the surface of the target region.
Higher-Order Ambisonics (HOA) encoding from sparse, irregular microphone arrays remains a critical challenge for consumer spatial audio capture in immersive communication and XR. We propose Flow-HOA, a generative framework that jointly optimizes a multi-dimensional objective encompassing time-domain, spectral, and spatial fidelity while producing a deployable, time-invariant bank of Finite Impulse Response (FIR) encoding filters. Using conditional flow matching, the model learns to map a simple prior distribution to the target distribution of FIR filter coefficients. Training is guided by a composite loss that balances time-domain waveform fidelity, multi-resolution spectral consistency, sub-band energy preservation, and spatial directivity constraints. Objective evaluations on synthetically simulated data demonstrate improved performance over strong model-based baselines in both signal fidelity and spatial accuracy metrics. Subjective listening tests on real microphone array recordings further confirm that Flow-HOA yields higher overall sound quality with reduced artifacts, demonstrating generalization from synthetic training data to real-world capture conditions.
Yuhuan You, Yufan Qian, Tianshu Qu +2
State Key Laboratory of General Artificial Intelligence, School of Intelligence Science and Technology, Peking University, Beijing, China · 2Beijing Xiaomi Mobile Software Co., Ltd · 3Xiaomi Communications Co., Ltd
Driven by the rapid growth of immersive teleconferencing and generative spatial audio, efficient low-bitrate coding of first-order Ambisonics (FOA) has become increasingly important. In this work, we develop a lightweight parametric FOA codec that retains the standard Directional Audio Coding analysis and synthesis while redesigning the spatial metadata quantization scheme. Rather than quantizing direction-of-arrival (DOA) and diffuseness independently, we combine them into a 3-D directivity vector and jointly quantize these vectors across frequency bands via residual vector quantization (RVQ). The RVQ codebooks are optimized within minutes via stage-wise k-means without backpropagation, yielding a constant-bitrate (CBR) representation that decouples metadata rate from the number of frequency bands. Evaluations show that our approach outperforms low-bitrate perceptual codecs in FOA reconstruction, remains competitive with neural codecs on the downstream Sound Event Localization and Detection task, and maintains robust performance when paired with different external monaural codecs. Given its lightweight training, strong performance, and CBR design, we consider the proposed method to be a favorable and reproducible baseline for future FOA codec research.
Wei-Ting Lai, Amy Bastine, Lachlan Birnie +2
Audio & Acoustic Signal Processing Group, The Australian National University, Canberra, Australia
This paper presents virtual upmixing of steering vectors captured by a fewer-channel spherical microphone array. This challenge has conventionally been addressed by recovering the directions and signals of sound sources from first-order ambisonics (FOA) data, and then rendering the higher-order ambisonics (HOA) data using a physics-based acoustic simulator. This approach, however, struggles to handle the mutual dependency between the spatial directivity of source estimation and the spatial resolution of FOA ambisonics data. Our method, named SIRUP, employs a latent diffusion model architecture. Specifically, a variational autoencoder (VAE) is used to learn a compact encoding of the HOA data in a latent space and a diffusion model is then trained to generate the HOA embeddings, conditioned by the FOA data. Experimental results showed that SIRUP achieved a significant improvement compared to FOA systems for steering vector upmixing, source localization, and speech denoising.
Emilio Picard, Diego Di Carlo, Aditya Arie Nugraha +2
RIKEN AIP, UP1 EMS · Center for Advanced Intelligence Project, RIKEN, Japan · Sorbonne University, France +5