Spherical Harmonic Coefficient Transformation for Audio Bitstream Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Higher-order Ambisonics (HOA) representations require a large number of coefficients for real-time transmission of audio signals, making them impractical due to high data rates, and existing compression methods are inefficient as they compress in the spatial domain rather than the HOA domain.
Innovation Solution
The method involves transforming spherical harmonic coefficients using linear invertible transforms like rotation, translation, DCT, DFT, or vector-based decompositions such as SVD, PCA, and KLT to reduce the number of coefficients needed for representation, specifying transformation information in the bitstream, and only including non-zero coefficients above a threshold value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If HOA representation is used to represent sound field independently of speaker geometry, then adaptability is improved, but data rate increases making real-time transmission impractical
Solution Approach 1:
The patent extracts and transmits only the necessary HOA coefficients above a threshold value, removing redundant coefficients. This selective extraction maintains sound field representation quality while significantly reducing the data rate for real-time transmission.
Solution Approach 2:
The patent applies thresholding to change the parameter of coefficient significance, keeping only coefficients above a certain threshold. This parameter change reduces the quantity of transmitted data while preserving the essential sound field information needed for adaptability.
2Measurement precision
If all HOA coefficients are transmitted to maintain sound field accuracy, then measurement precision is improved, but loss of information increases due to inefficient compression
Solution Approach 1:
The patent changes the parameter of coefficient selection by applying a threshold, keeping only coefficients above this threshold. This maintains sufficient sound field representation accuracy while minimizing information loss through efficient compression of only relevant coefficients.
Solution Approach 2:
The patent transmits a partial set of coefficients (only those above threshold) rather than all coefficients. This partial action is sufficient to maintain sound field accuracy while reducing redundant data transmission and minimizing information loss.
3Productivity
If linear invertible transform is applied to reduce number of coefficients, then productivity is improved, but device complexity increases due to transformation operations
Solution Approach 1:
The patent applies linear invertible transforms as a preliminary step before coefficient selection and transmission. This preliminary transformation organizes the data in a way that facilitates subsequent thresholding and reduces the number of coefficients needing transmission, improving coding efficiency despite added transformation complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In general, techniques are described for transforming spherical harmonic coefficients. A device comprising one or more processors may perform the techniques. The processors may be configured to parse the bitstream to determine transformation information describing how the sound field was transformed to reduce a number of the plurality of hierarchical elements that provide information relevant in describing the sound field. The processors may further be configured to, when reproducing the sound field based on those of the plurality of hierarchical elements that provide information relevant in describing the sound field, transform the sound field based on the transformation information to reverse the transformation performed to reduce the number of the plurality of hierarchical elements.