Spatial Audio Processing with Reference Head Orientations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Spatial audio data transmission, decoding, and rendering are complex and inefficient, particularly in split-rendering systems where multiple devices require individual processing, leading to increased power consumption and redundant processing.
Innovation Solution
A system that determines a reduced set of reference head orientations based on user head orientations from multiple devices, processes spatial audio signals for these orientations, and transmits associated metadata for local adjustment, reducing redundant processing and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If spatial audio signals are processed for each individual device with unique head orientations, then the accuracy of spatial audio rendering is improved, but the computational load and power consumption increase significantly
Solution Approach 1:
The patent merges the processing requirements of multiple devices by identifying that many devices have similar or identical head orientations. Instead of processing spatial audio signals separately for each device, the system groups devices with similar orientations and processes a single representative signal for the group, thereby reducing overall computational load and power consumption while maintaining rendering accuracy.
Solution Approach 2:
The patent applies partial action by processing spatial audio signals only for a subset of unique head orientations rather than for every individual device. By determining that a smaller number of reference head orientations are needed to represent the orientations of multiple devices, the system reduces processing to the essential minimum required for accurate rendering across all devices.
2Adaptability or versatility
If spatial audio signals are processed for each individual device, then each device receives customized audio rendering, but redundant processing occurs and efficiency decreases
Solution Approach 1:
The patent combines the processing tasks for multiple devices into a single processing operation by grouping devices with similar head orientations. This allows the system to generate one spatial audio signal that can be efficiently distributed to multiple devices, eliminating redundant processing while maintaining customized rendering for each device based on its specific orientation.
Solution Approach 2:
The patent reduces processing to the partial action of generating spatial audio signals only for the essential unique head orientations needed. By determining the minimum set of reference orientations required to represent all devices, the system achieves efficient processing without sacrificing the ability to provide customized rendering for each individual device.
3Adaptability or versatility
If a large number of spatial audio signals are generated for different head orientations, then more devices can be served simultaneously, but the complexity of the processing system increases
Solution Approach 1:
The patent merges multiple processing operations into a single streamlined process by determining a reduced set of reference head orientations that can represent multiple devices. This approach allows the system to serve more devices simultaneously without increasing complexity, as the same processing pipeline is used for all devices, just with fewer reference signals to generate.
Solution Approach 2:
The patent applies partial action by generating spatial audio signals only for the minimum necessary number of reference head orientations. By carefully selecting this partial set to represent all possible device orientations, the system achieves the ability to serve more devices without increasing processing system complexity.
Data Source
AI summary
Example embodiments relating to audio signal processing are disclosed. For example, a method may comprise receiving an audio signal, and receiving, from a plurality M of devices, respective requests for a spatial audio signal, wherein the respective requests indicate respective user head orientations. The method may further comprise determining a plurality N of reference head orientations, wherein N<M and processing the received audio signal to obtain a plurality of spatial audio signals respectively associated with the plurality N of reference head orientations and a plurality of sets of metadata respectively associated with the plurality of spatial audio signals. The method may further comprise transmitting, to at least one of the devices, a selected spatial audio signal and associated metadata, wherein the selection is based on which reference head orientation is most closely associated with the user head orientation indicated in the request from the at least one device.