Multi-Microphone Wind Noise Suppression Using Signal Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wind noise, being impulsive and non-stationary, degrades the quality and intelligibility of speech signals picked up by microphones, as conventional noise suppression methods fail to effectively attenuate it due to its bursty nature and varying amplitude.
Innovation Solution
A multi-microphone system with spatially separated primary and reference microphones detects and suppresses wind noise by utilizing time-varying blocking matrices and active noise cancelers, exploiting the difference in speech signal magnitude between the two microphones to filter out wind noise, and employing correlation-based and spectral derivation-based methods for accurate detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional noise suppression schemes are used to attenuate noise, then acoustic noise and system-introduced noise are suppressed, but wind noise is not effectively attenuated due to its non-stationary and impulsive nature
Solution Approach 1:
The patent divides the noise suppression task into two separate stages: first detecting wind noise using multiple microphones and correlation analysis, then applying targeted suppression only when wind noise is detected. This segmentation allows the system to handle wind noise differently from other noise types, resolving the contradiction by making the suppression effective specifically for impulsive non-stationary wind noise while maintaining reliability for other noise conditions.
Solution Approach 2:
The patent introduces a correlation-based detection mechanism as an intermediary between noise detection and suppression. By computing correlation between signals from multiple microphones and comparing it to a threshold, the system creates an intermediate detection stage that identifies wind noise conditions before applying suppression, thereby making the overall system reliable for handling wind noise's unique characteristics.
2Measurement precision
If multiple microphones are used to detect wind noise, then wind noise detection accuracy is improved, but device complexity increases
Solution Approach 1:
The patent uses multiple microphones arranged in space to detect wind noise, adding a spatial dimension to the detection process. By analyzing correlations between signals from different spatial positions and comparing against temporal thresholds, the system achieves accurate wind noise detection without requiring complex individual microphone structures, thus improving measurement precision while managing device complexity through spatial rather than structural complexity.
Data Source
AI summary
Unlike sound based pressure waves that go everywhere, air turbulence caused by wind is usually a fairly local event. Therefore, in a system that utilizes two or more spatially separated microphones to pick up sound signals (e.g., speech), wind noise picked up by one of the microphones often will not be picked up (or at least not to the same extent) by the other microphone(s). Embodiments of methods and apparatuses that utilize this fact and others to effectively detect and suppress wind noise using multiple microphones that are spatially separated are described.