Earphone Call Noise Reduction with Energy-Based Voice Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing call noise reduction methods using AI models fail to effectively suppress wind noise, leading to low intelligibility of voice signals in high wind environments.

Innovation Solution

A call noise reduction method for earphones that utilizes multiple microphones to determine the presence of wind noise and adjusts the target voice signal based on energy differences and coherence data, followed by AI noise reduction processing to enhance wind noise suppression.

Engineering Contradictions & Design Principles

VSEngineering 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 intelligibility of voice signals deteriorates in high wind noise environments

Engineering Contradiction:
Improvewind noise suppressionVSAvoidvoice signal intelligibility
Core Design Contradiction:
Object-affected harmful factorsVSMeasurement precision

Solution Approach 1:

The patent divides the voice signal processing into multiple independent paths: a first path processes signals from a first microphone, a second path processes signals from a second microphone, and a third path processes signals from a third microphone. Each path independently analyzes coherence data and energy levels to identify and remove wind noise components, thereby improving voice intelligibility while maintaining noise suppression capability

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces coherence data as an intermediary metric to detect wind noise characteristics. By calculating coherence between signals from different microphones and comparing it against threshold values, the system can identify wind noise presence and selectively apply noise reduction only to affected frequency bands, preserving voice signal quality

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If multiple microphones are used to detect wind noise and select target voice signals, then voice signal intelligibility is improved, but device complexity increases

Engineering Contradiction:
Improvevoice signal intelligibilityVSAvoidmicrophone system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements dynamic switching between different microphone configurations based on real-time coherence analysis. The system can adaptively select which microphone signals to use as target voice signals depending on wind noise conditions detected through coherence calculations, optimizing performance without requiring all microphones to operate simultaneously at full processing capacity

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent performs preliminary coherence analysis and energy comparison before final noise reduction processing. By pre-identifying wind noise affected frequency bands through coherence threshold comparison and selecting appropriate target voice signals in advance, the system simplifies the subsequent noise reduction step and avoids complex real-time processing of all signal paths

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4632738A1Call noise reduction method and earphone
Publication Date: 2025.10.15 ANKER INNOVATIONS TECH CO LTD
  • EP4632738A1 patent drawingFigure 1
  • EP4632738A1 patent drawingFigure 2
  • EP4632738A1 patent drawingFigure 3~4

AI summary

The present application relates to a call noise reduction method and an earphone. The method comprises: in the presence of wind noise in an external environment, obtaining energy of a first voice signal collected by the first microphone, and obtaining energy of a second voice signal collected by the second microphone; if the difference between the energy of the first voice signal and the energy of the second voice signal is greater than a preset threshold, determining the second voice signal as a target voice signal; if the difference between the energy of the first voice signal and the energy of the second voice signal is less than or equal to the preset threshold, determining the target voice signal according to the first voice signal; and performing wind noise reduction processing on the target voice signal. The method can improve the intelligibility of voice signals. (Fig. 1)