First-Order Ambisonics (FOA) offers a practical representation for spatial audio, but its low-order reproduction introduces distinct perceptual limitations during binaural rendering. While recent methods synthesize Higher-Order Ambisonics (HOA) from FOA to overcome these constraints, the importance and potential of early reflections in the upscaling process have not yet been comprehensively explored. To bridge this knowledge gap and understand the potential of incorporating early reflections in FOA upscaling, this work develops an experimental framework that selectively upscales these reflections. By utilizing an energy-based criterion, the proposed framework identifies and selectively renders the strongest early reflections with high-order resolution. The selection mechanism is evaluated across three domains: long-term temporal, short-time temporal, and the joint time-frequency (TF) domain. Validation through objective metrics, including the Scale-Invariant Signal-to-Distortion Ratio (SI-SDR), the clarity index (C50), and Interaural Cross-Correlation (IACC), alongside listening tests, demonstrates that the TF-domain provides superior performance. By exploiting the sparsity of speech, the TF approach maintains robust performance in complex multi-talker scenarios, providing insights into the benefits of early reflections for the design of future parametric and data-driven FOA upscaling methods.
Figures & tables
Parameter
Value / Description
Number of Rooms
6 with volumes 72 – 1800 m 3
Rooms Dimensions ( L×W×H )
[6.0,20.0]×[4.0,15.0]×[3.0,6.0] m
T60 Range
[0.3,0.8] s
Source Positions
3 positions per room ( ≥0.5 m from boundaries)
Array Positions
10 varying distances per source position ( 30 total per room)
Direct-to-Reverberant Ratio
[−10,15] dB
Table 1: Details of the simulated acoustic environments and ranges of parameters.
Figure 1: Global sdr performance evaluated across all acoustic scenes.
Figure 2: sdr evaluating the reconstruction of the direct signal components.
Figure 3: sdr evaluating the reconstruction of direct and early reflection components.
Figure 4: Absolute error in clarity index ( ΔC50 ) across different reflection orders.
Figure 5: Error in iacc ( Δ iacc ) across different reflection orders.
Figure 6: Global sdr performance evaluated across two distinct DRR conditions. Top: DRR ∈ [-10,5] dB. Bottom: DRR ∈ [5,15] dB.
Figure 7: Performance of direct sound extraction ( sdr Direct) under multi-source scenarios (1 to 4 concurrent active speakers).
Figure 8: Global sdr under multi-source scenarios (1 to 4 concurrent active speakers).
Figure 9: Improvement of tf over lt in global sdr as a function of the drr , for single-source scenes (average of 50 scenes) and multi-source scenes (average of 130 scenes).
Screen
Anchor
lt
st
tf
1: Single talker, r=1
4.25
16.41
16.43
16.64
2: Two talkers, r=1
2.77
4.69
10.05
15.65
3: Two talkers, r=3
2.77
16.61
12.32
19.35
Table 2: sdr (dB) of the listening-test signals relative to the reference.
Figure 10: Box plots of the listening-test scores (10 participants) for the hidden reference (Ref), the foa anchor, and the lt , st and tf methods, in the three mushra screens. Boxes span the 25th–75th percentiles, horizontal lines indicate the median, whiskers extend to the most extreme non-outlying scores, and open circles indicate outliers.
Spatial audio enhances immersion by reproducing 3D sound fields, with Ambisonics offering a scalable format for this purpose. While first-order Ambisonics (FOA) notably facilitates hardware-efficient acquisition and storage of sound fields as compared to high-order Ambisonics (HOA), its low spatial resolution limits realism, highlighting the need for Ambisonics upscaling (AU) as an approach for increasing the order of Ambisonics signals. In this work we propose DiffAU, a cascaded AU method that leverages recent developments in diffusion models combined with novel adaptation to spatial audio to generate 3rd order Ambisonics from FOA. By learning data distributions, DiffAU provides a principled approach that rapidly and reliably reproduces HOA in various settings. Experiments in anechoic conditions with multiple speakers, show strong objective and perceptual performance.
Amit Milstein, Nir Shlezinger, Boaz Rafaely
ECE School Ben-Gurion University of the Negev, Be’er-Sheva, Israel
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