Binaural Decoder Gain Correction for Non-Energy-Conserving Upmix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for binaural decoding of multi-channel audio signals, particularly those using HRTF filtering, face challenges in energy conservation and spectral coloring due to non-energy-conserving upmixing rules, especially when dealing with downmixed signals and spatial parameters that are not calculated to reflect energy distribution.
Innovation Solution
A multi-channel decoder is developed that calculates a gain factor to reduce or eliminate energy errors by using upmix rule information and HRTF filter characteristics, allowing for energy-corrected binaural signal generation without fully rendering the multi-channel signal, and applies this gain factor during filtering to mitigate spectral coloring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If non-energy-conserving upmixing rules are used for binaural decoding, then computational efficiency is improved, but energy errors and spectral coloring artifacts occur
Solution Approach 1:
The patent modifies the upmixing process by introducing energy compensation parameters (gain factors) that adjust the energy distribution in the upmixed channels. These parameters are calculated based on the difference between the energy of the downmixed signal and the sum of energies of upmixed channels, thereby compensating for energy losses while maintaining computational efficiency.
Solution Approach 2:
The patent introduces an intermediary energy compensation stage between the non-energy-conserving upmixing process and the final binaural decoding. This intermediary calculates and applies gain factors to correct energy errors, acting as a mediator that preserves the computational benefits of non-energy-conserving rules while eliminating their harmful energy loss effects.
2Device complexity
If non-energy-conserving upmixing rules are used, then device complexity is reduced, but spectral coloring artifacts increase
Solution Approach 1:
The patent changes the parameters of the upmixing process by introducing energy compensation gain factors that are applied to the upmixed channels. These gain factors are calculated to match the energy of the downmixed signal, thereby preventing the spectral coloring artifacts that would otherwise result from non-energy-conserving upmixing, while keeping the decoder complexity low.
Solution Approach 2:
The patent converts the harmful effect of non-energy-conserving upmixing into a benefit by calculating the energy difference and using it to determine compensation gain factors. The energy error that would cause spectral coloring is instead used as the basis for correcting the energy distribution, thereby eliminating the artifact while maintaining the simplified decoder structure.
3Reliability
If energy compensation is applied, then audio quality is improved, but computational load increases
Solution Approach 1:
The patent applies partial energy compensation by calculating gain factors only for the necessary upmixed channels and applying them selectively. This partial action approach improves audio quality by correcting the most significant energy errors while avoiding the excessive computational load that would result from a more comprehensive energy compensation scheme.
Solution Approach 2:
The energy compensation process is self-service in the sense that the gain factors are calculated automatically from the energy difference between the downmixed and upmixed signals, without requiring external intervention or complex optimization. This self-calculating approach improves audio quality while minimizing the additional computational load.
Data Source
AI summary
A multi-channel decoder for generating a binaural signal from a downmix signal using upmix rule information on an energy-error introducing upmix rule for calculating a gain factor based on the upmix rule information and characteristics of head related transfer function based filters corresponding to upmix channels. The one or more gain factors are used by a filter processor for filtering the downmix signal so that an energy corrected binaural signal having a left binaural channel and a right binaural channel is obtained.