Dual-Microphone Earphone Noise Processing for Adaptive ANC Modes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing headsets fail to effectively implement user-specific audio processing functions such as active noise control, ambient sound hear-through, and augment hearing, failing to adapt to varying user requirements and environmental conditions.

Innovation Solution

A headset with dual microphones and adaptive processing modes that determine and optimize auditory perception based on scene type, using filtering and mixing techniques to enhance or suppress sounds according to user preferences and environmental factors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If active noise reduction (ANR) is implemented in earphones, then noise cancellation capability is improved, but power consumption increases and causes battery drain

Engineering Contradiction:
Improvenoise cancellation capabilityVSAvoidpower consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent implements periodic action by switching between different noise processing modes (ANC, ENC, and off states) based on environmental conditions and user behavior patterns. The system periodically adjusts noise processing intensity and uses machine learning to predict optimal switching times, reducing continuous power consumption while maintaining effective noise cancellation when needed.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent applies dynamics by making the noise processing system adaptive and adjustable. The machine learning model dynamically optimizes noise processing parameters based on real-time feedback from sensors and user interactions. The system can dynamically switch between full ANC, enhanced ENC, and minimal processing modes, allowing power consumption to vary according to actual needs rather than operating at fixed high power levels.

Inventive Principle:
Principle #15Dynamics

2Reliability

If environmental noise cancellation (ENC) is implemented to reduce external noise during calls, then call quality is improved, but self-talk detection accuracy deteriorates due to noise interference

Engineering Contradiction:
Improvecall qualityVSAvoidself-talk detection accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary mechanism by using multiple microphones and signal processing layers to mediate between the noisy external environment and the self-talk detection algorithm. The system processes signals through multiple stages, using reference microphones to capture environmental noise separately, then subtracts this noise component before analyzing the remaining signal for self-talk detection, thereby improving detection accuracy despite ENC being active.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent applies segmentation by dividing the audio signal processing into distinct components: environmental noise capture, voice signal extraction, and self-talk detection. This segmentation allows the system to handle noise cancellation and self-talk detection as separate but coordinated processes, improving the ability to detect self-talk even when ENC is actively processing environmental noise.

Inventive Principle:
Principle #1Segmentation

3Reliability

If machine learning model continuously optimizes noise processing, then noise processing effect is improved, but computational load and power consumption increase

Engineering Contradiction:
Improvenoise processing effectVSAvoidcomputational power consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent implements periodic action in the machine learning optimization process by training and updating the noise processing model at specific intervals rather than continuously. The system uses offline training phases where the model is updated with new data, then applies these updated parameters periodically during operation. This approach maintains improved noise processing effects while avoiding the continuous computational burden of real-time model retraining.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent applies preliminary action by pre-training the machine learning model with extensive noise data before deployment, and by pre-calculating optimal processing parameters for different environmental conditions. This allows the model to make accurate noise processing decisions during operation without requiring intensive real-time computation, as much of the heavy lifting has already been done in advance.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4167590B1Earphone noise processing method and device, and earphone
Publication Date: 2026.05.20 HUAWEI TECH CO LTD
  • EP4167590B1 patent drawingFigure 1
  • EP4167590B1 patent drawingFigure 2
  • EP4167590B1 patent drawingFigure 3

AI summary

Embodiments of this application provide a headset noise processing method, an apparatus, and a headset, to implement the desired effect based on a user requirement. The headset has at least two functions of an ANC function, an HT function, or an AH function. The headset includes a first microphone and a second microphone. The first microphone is configured to collect a first signal. The first signal indicates a sound in a current external environment. The second microphone is configured to collect a second signal. The second signal indicates an ambient sound in an ear canal of a user wearing the headset. The headset can be a left earphone or a right earphone. Processing modes or processing strengths of the left earphone and the right earphone may be the same or different. The headset obtains a target mode, where the target mode may be determined based on a scene type of the current external environment; and obtains a second audio signal based on the target mode, the first signal, and the second signal. The target mode is determined based on the scene type of the external environment, so that auditory perception effect can be optimized for the user in real time.