Parametric Audio Encoding Using Dynamic Weight Estimates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-channel audio encoding and decoding systems face challenges in maintaining audio quality, particularly for out-of-phase signals, due to the suppression of non-aligned components during down-mixing, leading to reduced energy and artifacts, and require complex phase alignment methods that are not robust or efficient.
Innovation Solution
A decoder and encoder system that uses weighted combinations of channel signals with different amplitudes for each time-frequency interval, generating weight estimates from up-mix parametric data to recreate the original multi-channel audio signal, thereby mitigating out-of-phase issues and maintaining low complexity and data rate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If passive down-mixing (simple averaging) is used, then the system complexity is low and ease of operation is high, but acoustic energy is substantially reduced and out-of-phase signal components are completely lost
Solution Approach 1:
The patent applies dynamics by making the down-mixing weights dynamic rather than static. The weights are adjusted based on the measured inter-channel phase difference and correlation for each time-frequency tile, allowing the system to adapt to varying signal conditions and preserve energy for out-of-phase components when necessary
Solution Approach 2:
The patent changes the parameters of the down-mixing process by introducing variable weights that depend on phase difference and correlation parameters. This allows the system to optimize energy preservation across different signal conditions while maintaining simplicity in the basic down-mixing structure
2Manufacturing precision
If phase alignment is performed before summation, then audio quality is improved, but the system complexity increases and algorithmic delay is introduced
Solution Approach 1:
The patent extracts only the essential information needed for quality improvement (inter-channel phase difference and correlation parameters) and uses this extracted information to adjust down-mixing weights, avoiding the need for complex full-phase-alignment processing while maintaining audio quality
Solution Approach 2:
The patent applies partial phase alignment by using phase difference information only to adjust the weights of the down-mixing operation rather than performing complete phase alignment of all signal components, reducing complexity while maintaining essential quality improvements
3Manufacturing precision
If different weights with different amplitudes are used for time-frequency intervals, then out-of-phase signal components are preserved and audio quality is improved, but more parametric data needs to be transmitted
Solution Approach 1:
The patent makes the existing parametric data (inter-channel phase difference and correlation parameters) serve multiple functions: they are used both for spatial audio reconstruction and for determining optimal down-mixing weights, eliminating the need for separate weight data transmission
Solution Approach 2:
The patent merges the weight determination function with the existing parametric data processing, combining multiple functions (spatial reconstruction and weight optimization) into a unified approach that uses the same transmitted parameters for both purposes
Data Source
AI summary
An encoder for a multi-channel audio signal which comprises a down-mixer (201, 203, 205) for generating a down-mix as a combination of at least a first and second channel signal weighted by respectively a first and second weight with different amplitudes for at least some time-frequency intervals. Furthermore, a circuit (201, 203, 209) generates up-mix parametric data characterizing a relationship between the channel signals as well as characterizing the weights. A circuit generates weight estimates for the encoder weights from the up-mix parametric data; and comprises an up-mixer (407) which recreates the multi channel audio signal by up-mixing the down-mix in response to the up-mix parametric data, the first weight estimate and the second weight estimate. The up-mixing is dependent on the amplitude of at least one of the weight estimate(s).