Adaptive Beamformer for Hearing System Voice Pickup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In hearing systems with a separate microphone unit, the variable placement and orientation of the microphone relative to the user's mouth lead to a decrease in signal-to-noise ratio (SNR) due to the adaptive beamformer noise reduction system's reduced effectiveness, as it assumes a fixed position and direction of the target signal source.
Innovation Solution
An adaptive beamformer-noise reduction system is employed in the microphone unit, which estimates the noise power spectral density and updates the beamformer weights based on the user's voice activity and position, using multiple microphones to attenuate noise from other directions while preserving the target signal, and includes a voice activity detector to optimize noise reduction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a separate microphone unit is used with variable placement and orientation, then flexibility in placement is improved, but signal-to-noise ratio decreases due to reduced effectiveness of fixed-orientation beamformer
Solution Approach 1:
The patent applies the Dynamics principle by transitioning from a fixed-orientation beamformer to an adaptive beamformer that dynamically adjusts its orientation to track the user's mouth position. The system continuously estimates the direction of arrival of the user's voice and reorients the beamformer's sensitive direction accordingly, allowing the microphone unit to maintain optimal signal-to-noise ratio even when placed in variable positions or orientations on the user's body.
Solution Approach 2:
The patent implements the Feedback principle by using the detected user voice signal to continuously update the beamformer's orientation. The system monitors the direction of arrival of the target speech signal and feeds this information back to adjust the beamformer weights, creating a closed-loop system that automatically adapts to changing mouth positions and orientations without requiring manual recalibration.
2Reliability
If adaptive beamformer updates are performed frequently to track mouth position, then signal-to-noise ratio is improved, but computational complexity and processing time increase
Solution Approach 1:
The patent applies the Periodic action principle by performing beamformer updates at specific intervals triggered by voice activity detection rather than continuously. The system updates the beamformer orientation periodically when voice activity is detected, and remains dormant during silent periods, thereby reducing overall computational complexity while maintaining effective tracking of mouth position during speech.
Solution Approach 2:
The patent implements the Parameter changes principle by adapting the update frequency and computational intensity based on voice activity levels. During active speech, the system performs frequent updates with full computational effort, while during silent periods or low-activity states, it reduces update frequency and computational load, thereby optimizing the balance between signal-to-noise ratio maintenance and computational resource consumption.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the signal-to-noise ratio by adaptively adjusting to the changing position and orientation of the microphone unit, improving the quality of the user's voice signal transmission in noisy environments.
Implementation Method 1
a multi-input beamformer filtering unit operationally coupled to said multitude of input units IU i , i=1, ..., M, and configured to determine filter weights w(k,m) for providing a beamformed signal, wherein signal components from other directions than a direction of a target signal source are attenuated, whereas signal components from the direction of the target signal source are left un-attenuated or are attenuated less
Implementation Method 2
configured to provide corresponding electric input signals X i (k,m) in a time-frequency representation in a number of frequency bands and a number of time instances, k being a frequency band index, m being a time index
Data Source
Figure 1A~1B
Figure 2
Figure 3~4
AI summary
The application relates to a hearing system comprising a hearing device and a separate microphone unit adapted for picking up a voice of a user. The microphone unit comprises a) a multitude M of input units for picking up or receiving a signal representative of a sound from the environment, M bering ≥ 2; b) an adaptive multi-input unit noise reduction system for providing an estimate Ŝ of a target signal s comprising the user's voice, the multi-input unit noise reduction system comprises a multi-input beamformer filtering unit configured to determine filter weights w(k,m) for providing a beamformed signal, wherein signal components from other directions than a direction of a target signal source are attenuated, whereas signal components from the direction of the target signal source are left un-attenuated; and c) antenna and transceiver circuitry for transmitting said estimate Ŝ of the user's voice to another device. The hearing system facilitates communication between a wearer of a hearing device and another person via a telephone. The invention may e.g. be used in hearing aids in connection with handsfree telephone systems, mobile telephones, teleconferencing systems, etc.