Dual Microphone Speech Signal Processing for Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Bluetooth headsets struggle to effectively reduce noise during calls, as existing noise reduction methods using beamforming only process specific frequency bands, resulting in poor noise reduction performance.
Innovation Solution
A speech signal processing method that utilizes two microphones, one in the ear canal and one externally, to preprocess signals in different frequency bands, increasing the signal-to-noise ratio through noise reduction, amplitude adjustment, and correlation processing, and outputs a full-band low-noise speech signal by combining these signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If beamforming is used to combine two channels of speech signals, then noise reduction processing can be performed on speech signals in a specific frequency band range, but the noise reduction effect is poor because only specific included angle ranges are processed
Solution Approach 1:
The patent divides the speech signal processing into different frequency bands (first frequency band and second frequency band) and processes each band separately using different microphones and beamforming strategies. This segmentation allows optimized noise reduction for each frequency range while maintaining overall spectral coverage.
Solution Approach 2:
The patent transitions from processing only spatial dimension (included angle) to processing both spatial and frequency dimensions. By applying frequency band division and processing different bands with different microphones, the solution adds a frequency dimension to the noise reduction approach, enabling full-band coverage rather than limited to specific angles.
2Object-affected harmful factors
If only external speech collectors are used for noise reduction, then the frequency band coverage is wide, but the signal-to-noise ratio is low due to large environmental noise
Solution Approach 1:
The patent merges the advantages of two different microphone configurations: ear canal microphones (which provide high signal-to-noise ratio by isolating user speech) and external microphones (which provide wide frequency coverage). By combining signals from both types of microphones through frequency band division and beamforming, the system achieves both high signal quality and comprehensive frequency coverage.
Solution Approach 2:
Different microphones are assigned to different frequency bands based on their characteristics. Ear canal microphones are used for frequency bands where they excel (providing clean user speech), while external microphones handle other bands. This local optimization ensures each microphone type operates in its optimal performance range.
3Object-affected harmful factors
If ear canal speech collectors are used alone, then the signal-to-noise ratio is high, but the frequency band coverage is limited
Solution Approach 1:
The system makes the ear canal microphone system multi-functional by combining it with external microphones. The ear canal microphones primarily handle user speech capture with high signal-to-noise ratio, while external microphones supplement the frequency coverage. Through frequency band division and signal combining, the system achieves both high signal quality and wide frequency coverage, making the overall system universally applicable.
Data Source
AI summary
A speech signal processing method and apparatus. The method includes preprocessing a speech signal that is in a first frequency band and that is collected by an ear canal speech collector, to obtain a first speech signal; preprocessing a speech signal that is in a second frequency band and that is collected by at least one external speech collector, to obtain an external speech signal, where frequency ranges of the first frequency band and the second frequency band are different; performing correlation processing on the first speech signal and the external speech signal to obtain a second speech signal; and outputting a target speech signal, where the target speech signal includes the first speech signal and the second speech signal.