HRTF Decomposition for Personalized Audio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for personalizing binaural sound reproduction require lengthy and costly measurements in anechoic chambers, making it impractical to provide high-quality spatial audio experiences for a wide audience, as they depend on individual morphology and are difficult to generalize beyond a small number of subjects.
Innovation Solution
A method that decomposes HRTF transfer functions into independent components and spherical harmonics, allowing for compression and adaptation of filters based on morphological data, enabling rapid and non-intrusive personalization of audio systems without extensive measurements, using a combination of rotation and homothety transformations to generate personalized filter sets for new individuals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional HRTF measurement methods are used in anechoic chambers, then measurement precision and reliability are improved, but loss of time and device complexity increase significantly
Solution Approach 1:
The patent uses pre-measured HRTF data from reference individuals as templates, copying and adapting these measurements to target users through morphological matching rather than performing new measurements. This eliminates the time-consuming anechoic chamber measurement process while preserving the essential acoustic characteristics through template-based adaptation.
Solution Approach 2:
HRTF measurements are performed in advance on a set of reference individuals with different morphologies, and these pre-measured templates are stored for later adaptation. This preliminary action eliminates the need for time-consuming measurements during actual application, allowing rapid personalization through template selection and transformation.
2Manufacturing precision
If individualized HRTF measurements are performed for each user, then manufacturing precision of audio personalization is improved, but loss of time and ease of manufacture deteriorate
Solution Approach 1:
The patent transforms HRTF filters by applying parameter changes based on morphological differences between reference and target users. Specifically, it modifies frequency response parameters and temporal characteristics through mathematical transformations (frequency warping, time scaling) adapted to the user's auricle morphology, achieving personalization without new measurements.
Solution Approach 2:
A single set of pre-measured HRTF templates from reference individuals serves multiple functions: they are used for both analysis (comparing morphologies) and synthesis (generating personalized filters) across different target users. This universal template approach eliminates the need for separate measurement and processing pipelines for each user.
3Reliability
If complete HRTF filter sets are stored for each individual, then reliability of audio reproduction is improved, but quantity of substance (data storage requirements) increases
Solution Approach 1:
The patent segments the HRTF filtering process into two independent components: (1) a universal set of basis filters that capture the essential acoustic characteristics common to all users, and (2) user-specific transformation parameters derived from morphological matching. This segmentation allows storing compact representation data rather than complete filter sets for each user.
Solution Approach 2:
Instead of storing complete HRTF filter sets for each user, the system stores compact transformation parameters (morphological features, scaling factors, frequency warping parameters) that define how to adapt the universal basis filters to individual users. This parameter-based representation dramatically reduces storage requirements while maintaining personalization capability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for processing customized data representative of the directivity of an customized audio system, the method comprising the following steps: - obtaining (101), for each individual of an initial set of individuals, at least one customized suite of filters; - determining (103) N independent components common to the suites of filters obtained; - decomposing (104) each of the suites obtained into a first base constructed from the N independent components with a view to obtaining, for each suite of filters, a first suite of weighting coefficients; - decomposing (105) each first suite of weighting coefficients into a second base of P independent components, so as to obtain a second suite of weighting coefficients; - storing (106) each second suite of weighting coefficients obtained in association with an identifier of an individual from among the initial set of individuals.