Echo Suppression Controller Using Secondary Adaptive Filter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing acoustic echo cancellers face challenges in low echo return loss scenarios and high computational costs, leading to residual echoes in microphone signals due to adaptive filter divergence and poor performance in adapting to changing environments.
Innovation Solution
A controller for an echo suppressor that uses a secondary adaptive filter and coherence estimators to form echo estimates and decision parameters, activating the echo suppressor based on convergence states of the primary adaptive filter to effectively suppress residual echoes without interfering with near-end speech.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a primary adaptive filter is used to form echo estimates in an echo canceller, then echo cancellation performance is improved, but residual echo remains due to filter divergence and freezing during near-end speech
Solution Approach 1:
The system segments the echo cancellation task into two parts: a primary adaptive filter for general echo cancellation and a secondary adaptive filter specifically for removing residual echo. The secondary filter operates independently to address the limitations of the primary filter during near-end speech periods.
Solution Approach 2:
A decision parameter generator acts as an intermediary between the primary adaptive filter and the secondary adaptive filter. It monitors the primary filter's output and determines when residual echo is present, triggering the secondary filter only when necessary.
2Reliability
If an echo suppressor is activated to remove residual echo, then echo suppression performance is improved, but near-end speech may be clipped during double talk periods
Solution Approach 1:
The system uses feedback through decision parameters that continuously monitor the relationship between the primary error signal and primary echo estimate. This feedback mechanism enables intelligent control of the echo suppressor, activating it only when residual echo is detected and deactivating it during near-end speech to prevent clipping.
3Difficulty of detecting and measuring
If conventional near-end speech detectors are used to control echo suppressor activation, then near-end speech detection is achieved, but double talk clipping occurs during high echo relative to near-end speech
Solution Approach 1:
The decision parameter system dynamically adapts to varying echo conditions by adjusting its detection thresholds and parameters based on the current acoustic environment. This dynamic approach allows accurate detection of near-end speech even when echo levels are high, preventing false activation of the echo suppressor during double talk.
Data Source
AI summary
A controller for an echo suppressor configured to suppress a residual echo of a far-end signal included in a primary error signal, the controller adapted for operation with a primary adaptive filter configured to form a primary echo estimate of the far-end signal included in a microphone signal and an echo canceller configured to cancel that primary echo estimate from the microphone signal so as to form the primary error signal, the controller comprising: a secondary adaptive filter configured to form a secondary echo estimate of the far-end signal comprised in the microphone signal; and control logic operable in at least two modes selected in dependence on a convergence state of the primary adaptive filter, the control logic being configured to control activation of the echo suppressor in dependence one or more transient or steady state decision parameters.