Dual Channel Echo Suppressor for Residual Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional acoustic echo cancellation systems in electronic devices with loudspeakers and microphones, especially in portable devices, fail to effectively remove residual echoes due to non-linearity and vibrations, leading to disruptive noise-like signals that degrade voice recognition and call quality.
Innovation Solution
A dual channel echo suppressor is implemented, which performs echo cancellation followed by residual echo suppression using adaptive filtering and least mean square algorithms to calculate echo-to-speech ratio, adjusting spectral gain to reduce residual echoes while maintaining user speech quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional linear adaptive filtering is used for echo cancellation, then the system complexity is low, but the residual echo cannot be effectively removed due to non-linearity and vibrations
Solution Approach 1:
The echo cancellation system is segmented into multiple independent processing channels (first channel and second channel), each handling specific signal components. This segmentation allows complex non-linear processing to be distributed across separate channels, improving effectiveness while managing overall system complexity
Solution Approach 2:
A non-linear processing stage is introduced as an intermediary between the linear adaptive filtering stage and the final output. This intermediary component specifically addresses the non-linear distortions and vibrations that conventional linear filters cannot handle, thereby improving echo cancellation effectiveness
2Reliability
If non-linear processing is applied to reduce residual echo, then echo cancellation effectiveness improves, but user speech is also attenuated reducing voice quality
Solution Approach 1:
Different processing qualities are applied to different signal components: linear adaptive filtering is applied to the reference signal channel while non-linear processing is selectively applied to the error signal channel. This local differentiation allows effective echo reduction while preserving user speech characteristics
Solution Approach 2:
The system uses feedback from the error signal (which contains both residual echo and user speech) to continuously adjust the processing parameters. This feedback mechanism enables the system to distinguish between echo components needing suppression and speech components needing preservation, reducing speech attenuation
3Reliability
If aggressive echo suppression is applied, then residual echo is reduced, but the overall speech level decreases affecting recognition accuracy
Solution Approach 1:
The system dynamically changes processing parameters based on signal conditions. By adjusting the degree of non-linear processing applied to different time segments and frequency components, the system achieves effective echo suppression while maintaining adequate speech signal levels for recognition
Data Source
AI summary
A method includes obtaining, by a processor, an audio echo signal and an audio desired signal from an acoustic echo correction stage of an electronic device, and converting the echo signal and the desired signal to the frequency domain. The method further includes grouping, by the processor, frequency bin results of respective frequency domain converted echo and desired signals into respective echo and desired sub-bands. A sub-band suppressor gain is estimated based on an estimated sub-band energy for the echo and desired sub-bands. The method further includes modulating the frequency domain converted desired signal to compensate for residual echo, the modulating based, at least in part, on the estimated sub-band suppressor gain, and the modulating producing a compensated frequency domain converted echo signal. The method also includes converting the compensated frequency domain converted desired signal into time domain converted audio output signal.