Hearing Device Own Voice Detector Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hearing devices face challenges in distinguishing between wanted and unwanted voice signals, making it difficult to attenuate unwanted voices while enhancing wanted voices, especially in environments with multiple speech sources.
Innovation Solution
A hearing device equipped with an own voice detector and noise reduction system that uses multiple microphones and beamformers to identify and attenuate noise signal components, including unwanted voices, while maintaining and enhancing the target voice signal from the user, using techniques like MVDR and MWF beamformers and recursive noise covariance matrix updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If voice activity detection is used to identify speech signals, then speech enhancement is achieved, but unwanted voice signals from other sources cannot be distinguished from wanted voice signals
Solution Approach 1:
The patent segments the speech detection task by introducing an own-voice detector that specifically identifies the user's own voice separately from other speech sources. This segmentation allows the system to treat wanted voice (user's own voice) and unwanted voice (other sources) differently, resolving the contradiction between detecting speech accurately and distinguishing wanted from unwanted voices.
Solution Approach 2:
The patent introduces an intermediary component (own-voice detector) that acts as a mediator between the general voice activity detector and the noise reduction system. This intermediary provides additional information about the source of detected voices, enabling the system to selectively attenuate unwanted voices while preserving wanted voices.
2Object-affected harmful factors
If noise reduction is applied during all time segments, then unwanted noise is attenuated, but the user's own voice signal is also attenuated
Solution Approach 1:
The patent applies dynamic control to the noise reduction process by making it conditional on voice activity detection results. The noise reduction is applied selectively during time segments when unwanted noise is present and when it does not interfere with the user's own voice, rather than being applied statically to all time segments. This dynamic approach resolves the contradiction between noise attenuation and voice signal preservation.
Solution Approach 2:
The patent uses feedback from the own-voice detector to control the noise reduction process. The detector provides real-time information about the presence and source of voice signals, which feeds back to the noise reduction system to adjust its attenuation levels accordingly, preventing attenuation of the user's own voice while still reducing unwanted noise.
3Measurement precision
If additional detectors like cameras are used to identify noise sources, then noise source identification accuracy is improved, but device complexity increases
Solution Approach 1:
The patent makes the existing voice activity detector multi-functional by configuring it to perform both general voice detection and own-voice detection. This allows the single detector to identify different noise sources (other speakers, background speech) without requiring additional specialized detectors, resolving the contradiction between identification accuracy and device complexity.
Solution Approach 2:
The patent enables the existing voice activity detector to serve multiple purposes including identifying noise sources, rather than requiring separate dedicated detectors for each function. The detector essentially serves itself by analyzing voice signals to determine both presence and source, eliminating the need for additional cameras or sensors.
Data Source
AI summary
A hearing device adapted for being located at or in an ear of a user, or for being fully or partially implanted in the head of a user comprises a) an input unit for providing at least one electric input signal representing sound in an environment of the user, said electric input signal comprising a target speech signal from a target sound source and additional signal components, termed noise signal components, from one or more other sound sources, b) a noise reduction system for providing an estimate of said target speech signal, wherein said noise signal components are at least partially attenuated, and c) an own voice detector for repeatedly estimating whether or not, or with what probability, said at least one electric input signal, or a signal derived therefrom, comprises speech originating from the voice of the user. The noise signal components are identified during time segments wherein the own voice detector indicates that the at least one electric input signal, or a signal derived therefrom, originates from the voice of the user, or originates from the voice of the user with a probability above an own voice presence probability (OVPP) threshold value. A method of operating a hearing device is further disclosed.