Ambisonic Gain Calculation for Low-Load 3D Audio Rendering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing MPEG-H Part 3:3D audio standard generates 19 spread audio object signals for each audio object, significantly increasing calculation loads during rendering processing.
Innovation Solution
A signal processing apparatus and method that directly finds ambisonic gain based on spread information, using an ambisonic gain calculation unit to perform rotation processing on reference position ambisonic gains, thereby reducing the need to generate 19 spread audio objects, and utilizing a gain table for interpolation to determine ambisonic gains for specific spread angles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If 19 spread audio object signals are generated for one audio object based on spread information, then the audio object size can be expressed, but the calculation load increases remarkably
Solution Approach 1:
The patent extracts only the essential spread information (spread angle and azimuth angle) from the audio object metadata, and uses this extracted information to directly calculate ambisonic gains through spherical harmonic functions, eliminating the need to generate all 19 spread audio object signals while preserving the audio object size expression capability
Solution Approach 2:
The patent replaces the mechanical process of generating 19 separate spread audio object signals with a mathematical calculation system that directly computes ambisonic gains using spherical harmonic functions and rotation matrices, substituting signal generation with efficient mathematical transformation
2Productivity
If rotation processing is performed on reference position ambisonic gain to find ambisonic gain based on object position information, then the calculation load is reduced, but the processing complexity increases
Solution Approach 1:
The patent performs preliminary calculation of reference position ambisonic gains (when the audio object is at azimuth angle 0 degrees) using spherical harmonic functions, and stores these reference values. During actual rendering, only rotation processing is needed to transform these pre-calculated reference gains to the actual object position, significantly reducing the calculation load
Solution Approach 2:
The patent transforms the problem from generating multiple discrete spread audio object signals in the spatial domain to calculating continuous ambisonic gains through mathematical functions in the frequency domain, using spherical harmonic expansion and rotation matrices to handle position transformations
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present technology relates to a signal processing apparatus and method capable of reducing calculation loads, as well as a program. A signal processing apparatus includes an ambisonic gain calculation unit configured to find, on the basis of spread information of an object, an ambisonic gain while the object is present at a predetermined position. The present technology is applicable to an encoder and a decoder.