Directivity Hearing Aid Sound Source Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In noisy environments, existing directivity hearing-aid devices struggle to effectively capture and distinguish sounds from specific objects due to interference from high-decibel sounds and physical constriction methods, which can introduce noise and affect sound reception.
Innovation Solution
A directivity hearing-aid device that uses an audio pick-up device, processor, selector, and speaker to capture sound within a beam range, employ deep learning algorithms to extract and amplify specific sound components, suppress other sounds, and adjust frequencies to enhance the detection of targeted sounds, while using a near-end filter to manage user voice interference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical constriction method is applied on the audio pick-up device to focus on receiving sound in the pick-up direction, then the sound reception directionality is improved, but the device easily makes noise due to speedy wind and cannot effectively distinguish sound from specific object when multiple high-decibel sounds are received
Solution Approach 1:
The patent replaces the mechanical physical constriction method with an electronic signal processing system. The audio pick-up device captures sounds in all directions, and the processor uses algorithms to identify and extract sounds from specific directions and objects, substituting mechanical directionality control with electronic sound source separation and identification techniques.
Solution Approach 2:
The patent changes the approach from fixed physical constriction to dynamic parameter adjustment in signal processing. The processor analyzes sound signals in real-time, adjusting identification parameters such as sound source location, frequency characteristics, and decibel levels to effectively distinguish target sounds from background noise and wind interference.
2Measurement precision
If physical constriction method is applied to focus sound reception, then directionality is improved, but the device cannot effectively distinguish and listen to sound from specific object when a plurality of high-decibel sounds are received in the pick-up direction
Solution Approach 1:
The patent segments the sound signal processing into multiple independent analysis stages. The processor divides the complex sound environment into individual sound sources, analyzing each sound's characteristics separately. This segmentation allows the system to identify and extract sounds from specific objects even when multiple high-decibel sounds are present in the pick-up direction.
Solution Approach 2:
The patent introduces an intermediary sound source identification system between the audio pick-up device and the user. This intermediary processor acts as a mediator that analyzes, identifies, and separates sounds from specific objects, translating the complex mixed sound signals into distinct, recognizable sound sources for the user.
3Measurement precision
If deep learning algorithm is used to extract sound feature point audios, then sound component identification accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent implements a self-service deep learning system where the processor automatically learns and adapts to different sound environments and sound source characteristics. The deep learning algorithm continuously trains on incoming sound data, automatically improving its identification accuracy without requiring manual configuration or complex external control systems.
Solution Approach 2:
The patent designs a universal deep learning processor that can handle multiple sound identification tasks with a single system. The same processor architecture and algorithm framework can identify various types of sound sources (speech, music, environmental sounds) and adapt to different environments, reducing overall device complexity compared to having separate specialized systems for each function.
Data Source
AI summary
The present disclosure relates a directivity hearing-aid device which primarily includes an audio pick-up device, a processor, a selector, and a speaker disposed in a case. The audio pick-up device receives a sound in a range which is pointed by a beam, and the processor generates a detected sound component by amplifying a sound feature point audio in a sound component corresponding to a detecting target which is selected by the selector and generates adjusted sound components by suppressing or shielding the sound feature point audio in the other sound components. Then the processor combines the detected sound component and the adjusted sound components to generate an output sound signal. Accordingly, the directivity hearing-aid device of the present disclosure provides selectively receiving the detected sound component and suppressing or shielding the other impurity sounds to facilitate the user hears audio from a specific object thereby.