Voice Noise Suppression with Timed Post-Detection Muting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise suppression techniques, such as adaptive filters and residual echo suppression units, are insufficient in removing noise components from voice signals until they are optimized, and may excessively suppress signal components with small amplitudes.
Innovation Solution
A noise suppression device that includes an adaptive filter unit, a noise generation detection unit, and a period suppression unit, which detects noise generation timing and suppresses the noise component separately for a predetermined period before the adaptive filter is optimized, preventing excessive suppression of the voice signal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If an adaptive filter is used to suppress noise components, then noise suppression capability is improved, but suppression is insufficient until the adaptive filter is optimized
Solution Approach 1:
The system performs preliminary noise suppression using a fixed filter before the adaptive filter is fully optimized. This preliminary action provides immediate noise suppression capability while the adaptive filter continues to learn and optimize, resolving the contradiction by providing early suppression without waiting for full optimization.
Solution Approach 2:
The noise suppression process is divided into two segments: a fixed filter segment that provides immediate suppression and an adaptive filter segment that provides optimized suppression. This segmentation allows the system to provide noise suppression at different stages of adaptation, addressing the time delay issue.
2Reliability
If a residual echo suppression unit is provided to suppress residual noise, then noise suppression is improved, but signal components with small amplitude are suppressed together with noise
Solution Approach 1:
Different suppression strategies are applied to different time periods: a fixed filter is used during the initial period when noise characteristics are not yet learned, and the adaptive filter is used after optimization. This local differentiation prevents excessive suppression of voice signals while maintaining effective noise suppression.
Solution Approach 2:
The filter characteristics dynamically change from a fixed initial filter to an adaptive optimized filter over time. This dynamic transition allows the system to adapt its suppression strength, providing stronger suppression when needed and gentler suppression to preserve voice quality when the adaptive filter is fully optimized.
Data Source
AI summary
A noise suppression device includes: an adaptive filter unit that suppresses, using an adaptive filter, a noise component contained in a voice signal generated from a voice captured by a voice input unit to generate a corrected voice signal; a noise generation detection unit that detects timing of generation of the noise component in the voice signal; and a period suppression unit that suppresses the corrected voice signal during a predetermined period of time after the timing of the generation of the noise component.


