Ear-worn Device Speech Boost Circuitry for Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional hearing aids and ear-worn devices face limitations in improving signal-to-noise ratio (SNR) beyond a certain point, leading to reduced amplification in noisy environments due to discomfort from background noise.
Innovation Solution
The integration of neural network-based noise reduction with automatic volume increase, implemented through speech boost circuitry, enhances SNR significantly, allowing for greater optimal gain on the combined audio signal, particularly in environments with significant non-stationary noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional signal processing techniques are used to enhance audio signals, then some SNR improvement is achieved, but further amplification cannot be applied without amplifying noise as well
Solution Approach 1:
The patent extracts and removes noise components from the audio signal using neural network-based noise reduction circuitry before applying amplification. This allows the speech signal to be amplified without proportionally amplifying the noise, thereby resolving the contradiction between achieving SNR improvement and avoiding noise amplification.
Solution Approach 2:
The patent introduces an intermediary processing stage (neural network-based noise reduction) between the audio input and amplification stages. This intermediary component separates the speech signal from noise, enabling selective amplification of speech while suppressing noise, thus resolving the technical contradiction.
2Object-affected harmful factors
If overall amplification is reduced in noisy environments, then noise discomfort is reduced, but speech clarity is compromised
Solution Approach 1:
The patent extracts the speech signal from the noisy audio environment using neural network-based separation. This allows the system to amplify only the extracted speech component while leaving the noise component at its original lower level, thereby maintaining speech clarity without causing noise discomfort.
Solution Approach 2:
The patent applies different processing qualities to different components of the audio signal: high gain is applied locally to the speech component while low or no gain is applied to the noise component. This local differentiation resolves the contradiction by enhancing speech clarity in specific signal portions without amplifying noise that causes discomfort.
3Measurement precision
If neural network-based noise reduction is applied, then SNR improvement is significantly enhanced, but the optimal gain on combined signal must be increased to maintain equivalent loudness
Solution Approach 1:
The patent implements a feedback mechanism where the speech boost circuitry monitors the output of the noise reduction circuitry and dynamically adjusts the gain to maintain equivalent loudness. This feedback loop automatically compensates for the SNR improvement achieved by noise reduction, resolving the complexity of manual gain adjustment while preserving perceived speech levels.
Data Source
AI summary
An ear-worn device may include control circuitry configured to activate speech mode or traditional mode. Based on activation of speech mode, the control circuitry may be configured to control noise reduction circuitry to perform neural network-based noise reduction using neural network circuitry, and to control speech boost circuitry to automatically increase a volume of the audio signal at its output relative to the volume of the audio signal at its input, where the increase in the volume is in addition to any amplification applied by WDRC circuitry. Based on activation of traditional mode, the control circuitry may be configured to control the noise reduction circuitry to cease to perform the neural network-based noise reduction, and to control the speech boost circuitry to cease to automatically increase the volume of the audio signal at its output relative to the volume of the audio signal at its input.