HRTF Parameter Generation via Frequency Sub-band Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional Head-Related Transfer Function (HRTF) databases are large and computationally complex, making real-time processing of multiple sound sources challenging, especially in mobile applications, due to their high data storage requirements and complexity.

Innovation Solution

The method involves splitting frequency-domain HRTF signals into sub-bands and generating parameters based on statistical measures, such as root-mean-square values and phase angles, allowing for reduced data storage and processing complexity, enabling efficient interpolation and processing of multiple sound sources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional HRTF databases are used to achieve accurate spatial sound positioning, then sound positioning accuracy is improved, but data storage requirements and computational complexity increase

Engineering Contradiction:
Improvesound positioning accuracyVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The HRTF database is divided into multiple subsets, each representing different spatial regions or frequency ranges. Instead of processing the entire database, the system identifies and uses only the relevant subset corresponding to the current sound source direction, significantly reducing computational complexity while maintaining positioning accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transforms the HRTF representation from time-domain impulse responses to frequency-domain parameters (magnitude and phase). This parameter transformation enables more efficient storage and faster processing, as the complex temporal convolution operations are replaced with simpler frequency-domain multiplications and parameter lookups.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If conventional HRTF databases are used to achieve accurate spatial sound positioning, then sound positioning accuracy is improved, but data storage requirements increase

Engineering Contradiction:
Improvesound positioning accuracyVSAvoiddata storage requirements
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential parameters (magnitude and phase information) from the complete HRTF impulse responses. By storing these extracted parameters in a condensed format organized in subsets, the system achieves accurate sound positioning with significantly reduced data storage requirements compared to storing full impulse response databases.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The HRTF data is reorganized from a traditional time-domain impulse response format to a frequency-domain parameter representation with explicit magnitude and phase components. This dimensional transformation enables more compact storage while preserving the information needed for accurate spatial positioning.

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

3Measurement precision

If full HRTF processing is used to process multiple sound sources, then processing accuracy is maintained, but processing speed decreases

Engineering Contradiction:
Improveprocessing accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system segments the HRTF database into multiple subsets and selectively applies only the relevant subset for each sound source based on its direction. This segmentation allows parallel processing of multiple sound sources with reduced computational overhead, as each source uses a tailored, minimized parameter set rather than the complete database.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP1927264B1Method of and device for generating and processing parameters representing hrtfs
Publication Date: 2016.07.20 KONINKLIJKE PHILIPS NV
  • EP1927264B1 patent drawingFigure 1~6
  • EP1927264B1 patent drawingFigure 2~3
  • EP1927264B1 patent drawingFigure 4~5

AI summary

A method of generating parameters representing Head-Related Transfer Functions, the method comprising the steps of a) sampling with a sample length (n) a first time-domain HRTF impulse response signal using a sampling rate (fs) yielding a first time- discrete signal, b) transforming the first time-discrete signal to the frequency domain yielding a first frequency-domain signal, c) splitting the first frequency-domain signal into sub-bands, and d) generating a first parameter of the sub-bands based on a statistical measure of values of the sub-bands.