Wearable Voice Pickup With Adaptive Signal Mixing for Wind Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wearable audio devices struggle with effectively reducing wind noise and ambient noise while maintaining clear speech intelligibility, especially in challenging environmental conditions.
Innovation Solution
The wearable audio device employs multiple microphones and processors to form beamformed signals, compare energy levels, and dynamically mix or filter signals to enhance voice output, using adaptive noise reduction techniques like spectral subtraction and high-pass filtering based on wind and ambient noise conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If beamforming is used to enhance speech signal, then speech intelligibility is improved, but wind noise and ambient noise are not sufficiently reduced
Solution Approach 1:
The patent segments the audio signal processing into multiple independent stages: beamforming for speech enhancement, wind noise detection, ambient noise detection, and selective mixing. Each stage handles specific noise types independently, allowing precise control over speech intelligibility while targeting specific harmful factors without compromising the other.
Solution Approach 2:
The system dynamically adjusts the mixing between beamformed signal and original microphone signals based on real-time wind noise and ambient noise conditions. The processor selectively applies beamforming only when and where it improves speech intelligibility, while maintaining original signals in conditions where beamforming may introduce artifacts or insufficient noise reduction.
2Object-affected harmful factors
If aggressive noise reduction is applied, then wind and ambient noise are reduced, but speech clarity and naturalness deteriorate
Solution Approach 1:
The patent applies different processing qualities to different signal components: beamforming is applied selectively to enhance speech directions while preserving original quality in other directions, wind noise reduction is applied only to identified wind components, and ambient noise reduction is applied only to identified ambient noise components. This localized approach prevents over-processing that would degrade speech clarity.
Solution Approach 2:
The system continuously monitors the processed output for speech clarity and naturalness, using this feedback to adjust the degree of noise reduction applied. When speech clarity deteriorates, the system reduces the aggressiveness of noise reduction, maintaining a balance between noise removal and speech quality preservation.
3Measurement precision
If beamformed signal is always used, then speech enhancement is maximized, but popping and crackling artifacts increase
Solution Approach 1:
The system dynamically switches between beamformed signal and original microphone signal based on real-time conditions. When beamforming produces popping or crackling artifacts, the processor detects these degradations and selectively replaces affected portions with the original signal, maintaining speech enhancement while eliminating artifacts.
Solution Approach 2:
The patent creates a composite signal by combining beamformed signal components with original microphone signal components. This composite approach allows the system to benefit from beamforming's speech enhancement while using the original signal to compensate for beamforming artifacts, resulting in a hybrid output that maximizes enhancement while minimizing popping and crackling sounds.
Data Source
AI summary
A wearable two-way communication audio device includes a first microphone that provides a first microphone signal, a second microphone that provides a second microphone signal, and a third microphone that provides a third microphone signal. The device also includes one or more processors that are configured to process the first microphone signal and the second microphone signal to form a first beamformed signal. The one or more processors compare energy in the first beamformed signal to energy in the first microphone signal, and, if energy in the first beamformed signal exceeds energy in the first microphone signal, then the one or more processors mix the first microphone signal and the third microphone signal to provide a mixed signal. The one or more processors may also generate a voice output signal for transmission to a far end recipient using the mixed signal.