Wind Noise Reduction in Microphone Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Consumer electronic devices with multiple microphones face challenges in effectively reducing wind noise, which degrades signal quality and interferes with speech and audio recordings.
Innovation Solution
A method and device for wind noise reduction involving multiple microphones, where signals are split into sub-bands and processed to minimize wind noise through weighted mixing and recombination, optimizing spectral thresholds to preserve spatial cues and avoid audible artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If wind noise reduction is applied to all frequency bands, then wind noise is reduced, but spatial audio fidelity and natural sound quality deteriorate
Solution Approach 1:
The frequency spectrum is divided into multiple sub-bands (e.g., low, mid, high frequencies) so that wind noise reduction can be applied selectively to specific bands where wind noise predominates, while preserving natural sound characteristics in other bands. This segmentation allows differentiated processing strategies for different frequency ranges.
Solution Approach 2:
Different quality characteristics are applied to different frequency sub-bands. Wind noise reduction with stronger attenuation is applied to low-frequency sub-bands where wind noise is most prominent, while minimal or no processing is applied to high-frequency sub-bands to preserve spatial audio fidelity and natural sound quality.
2Object-affected harmful factors
If aggressive wind noise reduction is applied, then wind noise is minimized, but audible artifacts and signal distortion increase
Solution Approach 1:
Instead of applying maximum wind noise reduction across all frequencies, the system applies partial action by targeting only specific frequency sub-bands where wind noise is detected. This selective approach reduces wind noise effectively while avoiding excessive processing that would generate audible artifacts and signal distortion in other frequency ranges.
Solution Approach 2:
The system continuously monitors the audio signal and uses feedback to detect wind noise presence and characteristics. Based on this feedback, the wind noise reduction processing is dynamically adjusted or activated only when and where wind noise is detected, preventing unnecessary processing that would create artifacts in clean audio segments.
3Object-affected harmful factors
If multiple microphones are used for wind noise reduction, then wind noise is reduced more effectively, but device complexity increases
Solution Approach 1:
The multiple microphones serve multiple functions: they capture spatial audio information for normal operation and simultaneously provide the differential signals needed for wind noise reduction. The same microphone array used for high-quality audio recording also enables effective wind noise cancellation through computational methods, eliminating the need for separate dedicated wind noise microphones.
Data Source
AI summary
A method of wind noise reduction. Left side and right side microphone signals are obtained. In a first stage wind noise reduction is applied to a first sub-band of one of the signals, below a spectral threshold NA. In a second stage the wind noise reduced first side signal, and the second side signal, are both split into a sub-band below a threshold NB less than NA. The sub-band of the first side signal is mixed with the sub-band of the second side signal to produce an aggregate third sub-band signal having reduced wind noise, which is recombined with the respective sub-bands above the threshold NB to produce output first and second side signals.