Audio Processing Device Head Direction Rotation Matrix Ambisonics

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Ambisonics reproduction systems require a large number of speakers and have a narrow sound reproduction range, making them impractical for home use or large spaces like movie theaters, and combining Ambisonics with binaural reproduction techniques increases operational complexity and memory usage.

Innovation Solution

A sound processing apparatus that includes a head direction acquisition unit, rotation matrix generation unit, and head-related transfer function composition unit to efficiently generate headphone drive signals by selecting appropriate rotation matrices and composing input signals in the spherical harmonic domain, reducing the need for a large speaker array and minimizing memory usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a large number of speakers are used to increase spatial resolution of sound, then sound quality is improved, but device complexity and cost increase making it unrealistic for home use

Engineering Contradiction:
Improvespatial resolution of soundVSAvoidspeaker array configuration
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent creates a virtual copy of a multi-speaker Ambisonics system through computational processing. Instead of physically implementing a large speaker array, the invention uses a head-mounted device with microphones to capture spatial sound information and processes it to generate virtual speaker positions, effectively copying the acoustic field of a complex system with minimal physical components

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical acoustic system of multiple physical speakers with an electronic computational system. The head-mounted microphones capture sound, and digital signal processing algorithms generate the spatial audio effect, substituting physical acoustic radiation with electronic signal manipulation to achieve the same spatial resolution

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If Ambisonics and binaural reproduction techniques are combined, then sound reproduction capability is improved, but the amount of operation and memory usage increases

Engineering Contradiction:
Improvesound reproduction capabilityVSAvoidoperational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges Ambisonics spatial encoding with binaural reproduction techniques into a unified processing framework. The head-related transfer functions are applied directly to the Ambisonics decoded signals in a combined operation, eliminating the need for separate processing stages and reducing overall computational complexity while maintaining both spatial accuracy and binaural realism

Inventive Principle:
Principle #5Merging (Combining)

3Measurement precision

If a large number of speakers are used to reproduce sound space, then sound quality is improved, but the reproduction range becomes narrow making it difficult to provide desired effect to entire audience

Engineering Contradiction:
Improvesound space reproduction qualityVSAvoidreproduction range
Core Design Contradiction:
Measurement precisionVSArea of stationary object

Solution Approach 1:

The patent transitions from a fixed spatial reproduction system to a mobile personal reproduction system. By moving the reproduction point to the user's head through head-mounted devices, the system creates a personal sound field that moves with the user, effectively expanding the reproduction range from a fixed location to the entire space where users can move

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP3402223B1Audio processing device and method, and program
Publication Date: 2020.10.07 SONY GROUP CORP
  • EP3402223B1 patent drawingFigure 1
  • EP3402223B1 patent drawingFigure 2
  • EP3402223B1 patent drawingFigure 3

AI summary

The present technique relates to sound processing apparatus and method, and a program that can more efficiently reproduce sound. A sound processing apparatus includes: a head direction acquisition unit that acquires a head direction of a user listening to sound; a rotation matrix generation unit that selects two first rotation matrices on the basis of the head direction from a plurality of first rotation matrices for rotation in a first direction held in advance, selects one second rotation matrix on the basis of the head direction from a plurality of second rotation matrices for rotation in a second direction held in advance, and generates a third rotation matrix on the basis of the selected two first rotation matrices and the selected one second rotation matrix; and a head-related transfer function composition unit that composes an input signal in a spherical harmonic domain, a head-related transfer function in the spherical harmonic domain, and the third rotation matrix to generate a headphone drive signal in a time-frequency domain. The present technique can be applied to a sound processing apparatus.