Adaptive Window Weighting for Multichannel Audio Delay Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio coding devices inaccurately estimate inter-channel time differences due to excessive or insufficient smoothing of cross-correlation values, affecting the encoding and decoding quality of multi-channel audio signals.
Innovation Solution
A delay estimation method that uses an adaptive window function to weight cross-correlation coefficients based on delay track estimation values and smoothed inter-channel time difference information, adjusting the smoothing factor to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a uniform smoothing factor is used for all cross-correlation values, then the processing is simple, but some cross-correlation values are excessively smoothed while others are insufficiently smoothed, leading to inaccurate inter-channel time difference estimation
Solution Approach 1:
The patent applies different smoothing factors to different cross-correlation values based on their specific characteristics. Instead of using a uniform smoothing factor, the system dynamically adjusts the smoothing factor for each cross-correlation value according to its magnitude and position, ensuring that each value receives appropriate smoothing treatment rather than excessive or insufficient smoothing.
Solution Approach 2:
The patent introduces dynamic adjustment of smoothing factors based on the characteristics of cross-correlation values. The smoothing factor is not fixed but varies dynamically according to the specific cross-correlation value being processed, allowing the system to adapt to different situations and achieve optimal smoothing for each value.
2Stability of the object's composition
If excessive smoothing is applied to cross-correlation values, then the estimation is more stable, but the accuracy of inter-channel time difference estimation deteriorates
Solution Approach 1:
The patent changes the parameter of smoothing factor from a fixed uniform value to a dynamically adjusted value. By modifying the smoothing factor parameter based on the characteristics of each cross-correlation value, the system achieves the right balance between stability and accuracy, avoiding both excessive and insufficient smoothing.
Solution Approach 2:
The patent applies localized smoothing treatment to different cross-correlation values by using different smoothing factors for different regions or magnitudes of cross-correlation values. This ensures that each value receives the appropriate degree of smoothing for its specific characteristics.
3Measurement precision
If insufficient smoothing is applied to cross-correlation values, then the accuracy is maintained, but the estimation becomes unstable and inaccurate
Solution Approach 1:
The patent adjusts the smoothing factor parameter dynamically to achieve appropriate smoothing. By changing the smoothing factor based on the specific characteristics of each cross-correlation value, the system ensures sufficient smoothing for stability while avoiding excessive smoothing that would harm accuracy.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
This application discloses a delay estimation method and apparatus, and belongs to the audio processing field. The method includes: determining a cross-correlation coefficient of a multi-channel signal of a current frame; determining a delay track estimation value of the current frame based on buffered inter-channel time difference information of at least one past frame; determining an adaptive window function of the current frame; performing weighting on the cross-correlation coefficient based on the delay track estimation value of the current frame and the adaptive window function of the current frame, to obtain a weighted cross-correlation coefficient; and determining an inter-channel time difference of the current frame based on the weighted cross-correlation coefficient, so as to resolve a problem that the cross-correlation coefficient is excessively smoothed or insufficiently smoothed, thereby improving accuracy of estimating an inter-channel time difference.