Personalized Spatial Audio Calibration via Perceived Direction Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing binaural rendering systems for spatial audio struggle to provide personalized head-related transfer functions (HRTFs) for each listener, as obtaining personalized HRTFs using specialized equipment in an anechoic chamber is impractical for production devices.
Innovation Solution
A calibration process that prompts the listener to indicate the perceived direction of sound origin, allowing for the determination of a new HRTF based on the perceived direction, which can be used in place of the original HRTF, and optionally employs barycentric mapping to adjust the angular position.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If personalized HRTFs are obtained using specialized equipment in an anechoic chamber, then the accuracy of spatial audio rendering is improved, but the complexity and impracticality of the calibration process increases
Solution Approach 1:
The patent uses pre-measured HRTF data from a reference listener as a copy that can be applied to other listeners. Instead of requiring specialized anechoic chamber equipment for each user, the system copies and adapts existing HRTF measurements, significantly reducing calibration complexity while maintaining reasonable accuracy through subsequent personalization adjustments
Solution Approach 2:
The patent introduces simple calibration equipment (such as a single loudspeaker and basic audio interface) as an intermediary between the complex anechoic chamber measurement process and the end user. This intermediary enables practical calibration by translating complex HRTF concepts into simple directional sound localization tasks that users can perform with minimal equipment
2Adaptability or versatility
If personalized HRTFs are obtained using specialized equipment in an anechoic chamber, then the personalization of spatial audio is improved, but the ease of manufacture and deployment decreases
Solution Approach 1:
The system copies HRTF data from reference measurements and applies it to multiple users through software processing. This approach enables personalization to be manufactured and deployed easily across many devices without requiring physical anechoic chambers at each location, as the personalization data can be digitally copied and distributed
Solution Approach 2:
The patent replaces the mechanical/physical anechoic chamber system with a software-based HRTF processing system. By substituting complex physical measurement infrastructure with computational methods for HRTF personalization, the system achieves both personalization and ease of deployment through software updates rather than physical manufacturing
3Ease of operation
If generic HRTFs are used without personalization, then the ease of operation is improved, but the accuracy of sound direction perception deteriorates
Solution Approach 1:
The patent performs preliminary HRTF personalization through a calibration process that adapts generic HRTFs to individual users before actual spatial audio use. This preliminary action of measuring user-specific directional perception and adjusting HRTFs ensures both ease of operation (automatic calibration) and accuracy (personalized sound direction perception)
Solution Approach 2:
The system uses feedback from the user's actual sound direction perception during calibration to iteratively adjust and personalize the HRTFs. By incorporating feedback about where users actually perceive sounds as originating, the system refines the personalized HRTF parameters to match individual anatomical and perceptual characteristics, achieving both accuracy and ease of use
Data Source
AI summary
Techniques include providing a user with a set of loudspeakers on a sphere and playing back sound from the set of loudspeakers one at a time. For each loudspeaker, the user points in a direction at which the user perceives the sound. The HRTF corresponding to a new direction based on the perceived direction is used instead, or the original HRTF at the perceived direction is used. In some implementations, the new direction is obtained from the perceived direction via a barycentric mapping. In some implementations, a difference between the new direction and the perceived direction is equal to a difference between the perceived direction and the original direction.