Earphone Wind Noise Reduction Through Microphone Signal Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise reduction methods, particularly in high wind environments, fail to effectively distinguish wind noise from voice signals, leading to poor intelligibility of voice signals due to the uncorrelated nature of wind noise and the limitations of AI models in distinguishing between the two.
Innovation Solution
A noise reduction method for earphones that utilizes multiple microphones to determine energy differences and coherence data to identify wind noise, followed by selective use of voice signals and AI noise reduction models to process the target voice signal, thereby improving wind noise reduction and intelligibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If an AI model is used to cancel wind noise from a voice signal collected by a main microphone, then noise reduction is achieved, but the effect is poor in high wind noise environments resulting in low intelligibility of voice signals
Solution Approach 1:
The patent divides the voice signal processing into multiple independent channels, each with its own microphone and AI noise reduction model. The system segments the wind noise reduction task across multiple microphones (first microphone, second microphone, third microphone) and processes each channel separately before combining results, allowing better handling of high wind noise environments while maintaining voice signal intelligibility
Solution Approach 2:
The patent introduces a coherence calculation mechanism as an intermediary to determine whether wind noise is present. By calculating coherence between signals from different microphones, the system can identify wind noise conditions and selectively apply noise reduction processing only when necessary, thereby maintaining voice signal quality and intelligibility while reducing wind noise interference
2Object-affected harmful factors
If multiple microphones are used to detect wind noise and select target voice signals, then wind noise reduction effectiveness is improved, but device complexity increases
Solution Approach 1:
The patent performs preliminary coherence calculation and wind noise detection before selecting the target voice signal for noise reduction processing. By预先 determining whether wind noise is present through coherence analysis of multiple microphone signals, the system avoids unnecessary complex processing in normal conditions, thus improving wind noise reduction effectiveness while controlling device complexity
Solution Approach 2:
The patent applies different processing strategies to different signal channels based on local conditions. Each microphone channel is evaluated independently for wind noise characteristics, and only channels exhibiting wind noise patterns undergo AI noise reduction processing. This localized approach enhances overall wind noise reduction effectiveness while minimizing the complexity impact of using multiple microphones
Data Source
AI summary
The present application relates to noise reduction methods and earphones. The methods may be applied to telephone calls to reduce wind noise interference. A method comprises: determining energy of a first voice signal received by a first microphone, and determining energy of a second voice signal received by a second microphone. The method further comprises selecting, based on whether a difference between the energy of the first voice signal and the energy of the second voice signal is greater than a preset threshold, one of the first voice signal or the second voice signal. In addition, the method comprises performing wind noise reduction processing on the selected one of the first voice signal or the second voice signal.