Adaptive Shadow Filter for Diffuse Noise Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dual-microphone noise cancellation methods in speech encoders face challenges in low-frequency noise reduction, particularly in environments with diffuse noise fields, as they often cancel desired speech signals and require frequent re-convergence when speech is present, leading to suboptimal performance.
Innovation Solution
An adaptive shadow filter is used in conjunction with a diffuse-noise-field detector to differentiate between diffuse noise and speech, allowing the filter coefficients to be transferred for effective noise cancellation only when diffuse noise is detected, thereby avoiding speech cancellation and reducing the need for abrupt re-convergence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If a standard noise canceller is used to achieve high noise reduction at low frequencies, then noise cancellation performance is improved, but desired speech signals may be partially cancelled and filter variations increase
Solution Approach 1:
The patent applies preliminary action by continuously adapting the shadow filter coefficients during periods when diffuse noise is detected, so that when speech occurs the filter is already prepared and does not need to re-converge abruptly. This prevents speech cancellation while maintaining noise reduction capability.
Solution Approach 2:
The patent introduces a shadow filter as an intermediary component that separately adapts to diffuse noise characteristics without directly processing the speech signal. This intermediary filter allows the main noise canceller to focus on speech preservation while the shadow filter handles diffuse noise adaptation, resolving the contradiction between noise reduction and speech integrity.
2Object-affected harmful factors
If the adaptive filter continuously adapts to cancel noise, then noise cancellation performance is improved, but filter variations increase when speech is present
Solution Approach 1:
The patent applies dynamics by making the filter adaptation behavior conditional based on the detected noise field type. When diffuse noise is detected, the shadow filter adapts continuously; when speech is detected, adaptation is paused. This dynamic adjustment of adaptation behavior reduces unnecessary filter variations while maintaining effective noise cancellation.
Solution Approach 2:
The patent uses feedback from the diffuse-noise-field detector to control the adaptation process. The detector's output about the current acoustic environment feeds back to the shadow filter, enabling it to adapt only when appropriate conditions (diffuse noise presence) are met, thereby stabilizing filter weights during speech periods.
3Productivity
If the step size is adjusted based on SNR estimation to control convergence speed, then adaptation performance is improved, but complexity of the system increases
Solution Approach 1:
The patent extracts the adaptation control function from the main noise canceller by creating a separate shadow filter that independently adapts to diffuse noise. This extraction simplifies the overall system by removing the need for complex SNR-based step size adjustment logic, as the shadow filter's adaptation is controlled solely by the diffuse-noise-field detector.
Data Source
AI summary
The present invention relates to a method and arrangement for an improved noise canceller in a speech encoder. Sound signals are captured at a primary microphone in conjunction with a reference microphone. An adaptive shadow filter is adapted to the correlation between the signals captured at the primary and reference microphones. Further, a diffuse-noise-field detector is introduced which detects the presence of diffuse noise. When the diffuse-noise-field detector detects diffuse noise, the filter coefficients of the adapted shadow filter is used by a primary filter to cancel the diffuse noise at the signal captured by the primary microphone. Since the filter coefficients of the adapted shadow filter only is used for cancellation when diffuse noise is solely detected, cancellation of the speech signal is avoided.