DiffAU: Diffusion-Based Ambisonics Upscaling

2025-09-30

AI Summary

Low spatial resolution and insufficient 3D auditory immersion of first-order Ambisonics (FOA) hinder high-fidelity spatial audio reproduction. To address this, we propose DiffAU, a cascaded diffusion model specifically designed for Ambisonics upscaling. DiffAU operates at the data distribution level, integrating spherical harmonic priors and geometric constraints of spatial audio, and employs a multi-stage cascaded architecture to enable stable, efficient generation from FOA to third-order Ambisonics (HOA). Unlike end-to-end mapping approaches, DiffAU explicitly models the order-upscaling process stepwise. Evaluated on anechoic multi-speaker scenarios, DiffAU significantly outperforms existing methods: it achieves average improvements of 2.1 dB in SNR and 0.35 in PESQ, with a subjective Mean Opinion Score (MOS) of 4.21/5.0—demonstrating substantial advances in perceptual realism and spatial fidelity.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

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.

Problem

• Upscaling first-order Ambisonics to higher orders for improved spatial resolution
• Enhancing 3D audio realism through diffusion-based generative modeling
• Generating third-order Ambisonics from low-order signals using cascaded diffusion

Innovation

• Diffusion model for Ambisonics upscaling
• Cascaded method generates third-order Ambisonics
• Learns data distributions for reliable reproduction