VR Ambisonic Rotation for Low-Compute Spatial Audio Adaptation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current psychoacoustic decoders struggle with rotating spatial components and audio objects in the ambisonics domain, requiring computationally expensive and power-intensive domain translations.

Innovation Solution

A psychoacoustic decoder rotates spatial components based on rotation information from motion sensors, forming rotated spatial components and constructing ambisonic signals, reducing computational demand and improving coding efficiency by utilizing rotation information for phase difference quantization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If binaural audio is used to simulate 3D sound in virtual reality, then spatial audio perception is improved, but audio quality degrades at certain frequencies due to head-related transfer function filtering

Engineering Contradiction:
Improvespatial audio perceptionVSAvoidaudio quality
Core Design Contradiction:
Ease of manufactureVSManufacturing precision

Solution Approach 1:

An equalization filter is introduced as an intermediary component between the binaural audio processing and the output. This filter compensates for the frequency-dependent attenuation caused by the head-related transfer function, particularly restoring frequencies above 4kHz that would otherwise be lost. The equalization filter acts as a mediator that corrects the distortion introduced by the binaural processing while preserving the spatial audio effects.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If head-related transfer function filtering is applied to create 3D sound perception, then spatial localization is improved, but high frequency sounds above 4kHz are attenuated and lost

Engineering Contradiction:
Improvespatial localizationVSAvoidhigh frequency sound
Core Design Contradiction:
Measurement precisionVSLoss of substance

Solution Approach 1:

The equalization filter is designed and applied in advance to compensate for the expected frequency attenuation. By pre-calculating and applying the inverse of the head-related transfer function's frequency response, the system restores high frequency content before it is permanently lost. This preliminary equalization action ensures that high frequency sounds are preserved while maintaining the spatial localization benefits of HRTF filtering.

Inventive Principle:
Principle #10Preliminary action

3Device complexity

If conventional audio processing is used, then implementation is simple, but audio does not adapt to different virtual reality environments or user preferences

Engineering Contradiction:
Improveimplementation simplicityVSAvoidaudio environment adaptation
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The audio processing system is made dynamic by allowing the equalization filter characteristics to change based on different virtual reality environments and user preferences. The filter can be adjusted to match different acoustic environments, user hearing characteristics, or preferred audio styles. This dynamic adaptability enables the system to optimize audio output for various scenarios while building upon the relatively simple foundation of conventional binaural processing.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4062657B1Soundfield adaptation for virtual reality audio
Publication Date: 2026.05.06 QUALCOMM INC
  • EP4062657B1 patent drawingFigure 1A
  • EP4062657B1 patent drawingFigure 1B
  • EP4062657B1 patent drawingFigure 1C

AI summary

An example device includes a memory configured to store at least one spatial component and at least one audio source within a plurality of audio streams. The device also includes one or more processors coupled to the memory. The one or more processors are configured to receive, from motion sensors, rotation information. The one or more processors are configured to rotate the at least one spatial component based on the rotation information to form at least one rotated spatial component. The one or more processors are also configured to reconstruct ambisonic signals from the at least one rotated spatial component and the at least one audio source, wherein the at least one spatial component describes spatial characteristics associated with the at least one audio source in a spherical harmonic domain representation.