Variable Threshold Audio Signal Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio signal decomposition methods into background and foreground components often result in reduced audio quality due to inadequate separation techniques.
Innovation Solution
An apparatus and method that generate a time sequence of audio signal blocks, analyze block characteristics, and separate them into background and foreground components using a variability-based separation threshold, allowing for a non-binary separation and individual processing of signal parts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a fixed separation threshold is used to decompose audio signals into background and foreground components, then the decomposition process is simple and fast, but the audio quality is reduced due to inadequate separation
Solution Approach 1:
The patent applies dynamics by transitioning from a fixed separation threshold to a variable separation threshold that adapts to the local characteristics of the audio signal. The threshold is dynamically adjusted based on the short-term statistical properties (mean and standard deviation) of the signal blocks, allowing the decomposition process to respond to changing signal conditions and achieve better separation accuracy without excessive complexity
Solution Approach 2:
The patent implements parameter changes by modifying the separation threshold parameter from a constant value to a variable value that changes according to the signal's local statistics. The threshold is calculated as a function of the mean and standard deviation of signal blocks, enabling the system to adapt to different signal conditions and improve decomposition quality
2Measurement precision
If a variable separation threshold based on signal variability is used, then the separation accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent applies partial action by calculating the variable threshold only for relevant signal blocks rather than uniformly processing all blocks with the same complexity. The method computes local statistics (mean and standard deviation) selectively and applies the variable threshold primarily where signal variability indicates the need for adaptive separation, reducing overall computational load while maintaining accuracy where needed
3Reliability
If traditional decomposition methods are used, then the processing is computationally efficient, but the audio quality is reduced due to inadequate separation techniques
Solution Approach 1:
The patent implements local quality by applying different separation strategies to different parts of the audio signal based on their local characteristics. Signal blocks with high variability (indicative of foreground content) receive adaptive variable threshold separation, while blocks with low variability use simpler methods. This localized approach improves overall audio quality by tailoring the separation technique to the specific needs of each signal segment
Data Source
AI summary
An apparatus for decomposing an audio signal into a background component signal and a foreground component signal, has: a block generator for generating a time sequence of blocks of audio signal values; an audio signal analyzer for determining a characteristic of a current block of the audio signal and for determining a variability of the characteristic within a group of blocks having at least two blocks of the sequence of blocks; and a separator for separating the current block into a background portion and a foreground portion wherein the separator is configured to determine a separation threshold based on the variability and to separate the current block into the background component signal and the foreground component signal, when the characteristic of the current block is in a predetermined relation to the separation threshold.