Spatial Audio Capture With Breathing Noise Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio capture devices struggle with unbalanced acoustic and microphone signals when microphones are in close proximity to a user's face, amplifying non-speech sounds like breathing and nasal noises while diminishing distant sounds, leading to signal imbalance and undesirable artifacts.
Innovation Solution
An audio processing device applies attenuation algorithms to microphone signals, classifying near-field and far-field sounds, using multi-channel Wiener filters and beamforming techniques to reduce non-speech sounds while preserving spatial qualities and phase relationships, thereby mitigating user-based noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If microphones are placed in close proximity to the user's face, then speech capture is improved, but non-speech sounds (breathing, nasal noises) are amplified and distant sounds are diminished
Solution Approach 1:
The patent segments the audio signal processing into distinct components: near-field sound processing for speech capture and far-field sound processing for environmental audio. By separating the processing paths for different spatial zones, the system can optimize speech capture while independently controlling noise amplification and distant sound preservation
Solution Approach 2:
The patent applies different processing qualities to different spatial regions: aggressive noise attenuation is applied locally to near-field non-speech sounds, while far-field sounds receive minimal processing to preserve their natural characteristics. This localized quality adjustment resolves the contradiction by targeting only the harmful near-field noises without affecting distant sounds
2Object-generated harmful factors
If noise attenuation is applied to reduce non-speech sounds, then breathing and nasal noises are reduced, but spatial qualities and phase relationships may be degraded
Solution Approach 1:
The patent employs dynamic processing that adapts to the spatial characteristics of incoming sounds. The system dynamically adjusts attenuation levels based on the detected spatial position and type of sound, applying stronger attenuation to non-speech near-field sounds while maintaining spatial integrity for far-field sounds through adaptive spatial filtering
Solution Approach 2:
The patent changes key processing parameters based on spatial classification: attenuation strength, filter characteristics, and processing intensity are all dynamically adjusted according to whether the sound originates from near-field or far-field sources. This parameter adaptation allows aggressive noise reduction for speech while preserving spatial qualities for environmental sounds
3Object-generated harmful factors
If aggressive attenuation is applied to non-speech sounds, then breathing noises are reduced by 10 dB or more, but speech and distant sound clarity may be compromised
Solution Approach 1:
The patent introduces spatial classification as an intermediary mechanism that mediates between aggressive noise attenuation and preservation of desired sounds. By classifying sounds into near-field and far-field categories, the system can apply appropriate attenuation levels to each category, ensuring that breathing noises are reduced while speech and distant sounds maintain their clarity
Solution Approach 2:
The patent performs preliminary spatial classification and sound type identification before applying attenuation. By pre-identifying which sounds are non-speech near-field noises versus desired speech or far-field sounds, the system can apply aggressive attenuation to the correct targets while protecting desired audio content from excessive processing
Data Source
AI summary
A device may include microphones worn on a head of a user. The device may include a processor, configured to obtain microphone signals from the plurality of microphones. The processor may attenuate breathing sound from the user by processing the microphone signals, resulting in attenuated microphone signals. The processor may render one or more output audio channels based on the plurality of attenuated microphone signals.