Adaptive Time-Domain Filtering for Low-Delay Audio Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing signal processing techniques in audio devices introduce significant group delay, leading to issues like perceivable echo, reduced perceptual coherence, and increased resource requirements for echo or feedback cancellation, which affect sound quality and communication integrity.
Innovation Solution
The method involves passing an input signal through an adaptive time domain filter, transforming it into a transform domain, analyzing the signal to determine desired gains, and synthesizing a minimum phase time domain filter characteristic to reduce group delay, allowing for real-time adaptation and low-latency processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If block processing with FFT analysis and IFFT overlap-add synthesis is used to improve signal processing performance, then processing capability is enhanced, but group delay increases significantly
Solution Approach 1:
The patent divides the frequency spectrum into multiple sub-bands and processes each sub-band separately using parallel filter banks. This segmentation allows independent processing of frequency components without requiring large buffer blocks, thereby maintaining processing capability while reducing the group delay associated with long FFT blocks.
Solution Approach 2:
The patent transitions from time-domain block processing to a frequency-domain parallel processing architecture using filter banks. By transforming the processing dimension from sequential time blocks to parallel frequency sub-bands, the system achieves efficient signal processing with minimized phase distortion and reduced group delay.
2Reliability
If adaptive filtering is implemented to improve echo and feedback cancellation, then communication quality is enhanced, but computational resource requirements increase
Solution Approach 1:
The patent segments the adaptive filtering process into multiple parallel sub-band filters, each handling a specific frequency range. This segmentation distributes the computational load across multiple simpler filters rather than requiring one complex full-band adaptive filter, reducing overall computational resource requirements while maintaining echo and feedback cancellation effectiveness.
Solution Approach 2:
The patent applies adaptive filtering selectively to specific sub-bands where echo and feedback cancellation are most needed, rather than uniformly across the entire frequency spectrum. This partial action approach optimizes computational resource usage by concentrating processing effort where it provides maximum communication quality improvement.
3Productivity
If multi-band signal processing techniques are used to improve performance, then signal processing effectiveness is enhanced, but device complexity increases
Solution Approach 1:
The patent combines multiple filter banks and processing stages into an integrated multi-band processing architecture. By merging the analysis filter bank, sub-band processing units, and synthesis filter bank into a unified structure, the system achieves effective multi-band signal processing while managing device complexity through systematic integration and shared computational resources.
Data Source
AI summary
Adaptively processing an input signal, such as an input signal of a hearing aid. The input signal is passed through an adaptive time domain filter to produce an output signal. At least one of the input signal and the output signal is used as an analysis signal. The analysis signal is transformed into a transform domain to produce a transformed analysis signal, which is analyzed to produce a desired gain for each respective transform domain sub-band. A minimum phase time domain filter characteristic is synthesized which approaches the desired gains. The adaptive filter is updated with the synthesized filter characteristic.