Hearing Device Pitch Estimator Speech Intelligibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hearing aid users face degraded speech intelligibility in noisy multi-talker environments, as existing speech enhancement processing schemes are intrusive and require a reference speech signal, which is often unavailable, leading to sub-optimal estimation and application in monaural systems.
Innovation Solution
A hearing device and system that non-intrusively estimate speech intelligibility using a pitch estimator and direction of arrival, reconstructing a reference speech signal based on microphone input signals to control processing, allowing for adaptive speech enhancement without a clean speech signal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If speech enhancement processing schemes are applied in noisy multi-talker environments, then speech intelligibility may be improved in some listening environments, but the processing becomes intrusive and requires a reference speech signal that is often unavailable
Solution Approach 1:
The patent introduces a pitch estimator as an intermediary component that extracts pitch parameters from the noisy input signal without requiring a reference speech signal. This intermediary enables the speech intelligibility estimator to function in monaural systems by providing critical pitch information that would otherwise be unavailable, thus resolving the contradiction between assessment accuracy and system complexity
Solution Approach 2:
The system performs self-service by estimating speech intelligibility using only the microphone input signal and pitch estimation, without requiring external reference signals or complex processing schemes. The pitch estimator and speech intelligibility estimator work together to enable the hearing device to autonomously assess and adapt to listening environments
2Adaptability or versatility
If existing speech enhancement processing schemes are used, then speech intelligibility may be enhanced in certain environments, but the schemes are detrimental in other listening environments due to lack of adaptability
Solution Approach 1:
The patent implements dynamics by enabling the hearing device to continuously estimate speech intelligibility and adapt processing parameters in real-time based on the estimated values. The controller adjusts processing schemes dynamically according to the speech intelligibility indicator, allowing the system to adapt to varying listening environments while maintaining reliability
Solution Approach 2:
The system employs feedback by using the estimated speech intelligibility indicator to control the processing of the input signal. The speech intelligibility estimator provides continuous feedback to the controller, which adjusts the processing scheme accordingly, creating a closed-loop system that adapts to changing acoustic environments
3Measurement precision
If intrusive speech enhancement processing is applied, then speech clarity may be improved, but the processing requires a reference speech signal that is unavailable in monaural systems
Solution Approach 1:
The pitch estimator serves as an intermediary that bridges the gap between the noisy input signal and the speech intelligibility estimator. By extracting pitch parameters from the monaural input signal, it enables accurate speech intelligibility estimation without requiring a reference speech signal, thus maintaining ease of operation
Solution Approach 2:
The patent replaces the mechanical requirement of a reference speech signal with a computational approach using pitch estimation and speech intelligibility estimation algorithms. This substitution eliminates the need for physical reference signals while maintaining estimation accuracy through signal processing techniques
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure provides a hearing device comprising a set of microphones comprising a first microphone for provision of a first microphone input signal, and a processor for processing input signals and providing an electrical output signal based on input signals. The hearing device a receiver for converting the electrical output signal to an audio output signal, and a controller operatively connected to the set of microphones, the controller comprising a speech intelligibility estimator for estimating a speech intelligibility indicator indicative of speech intelligibility based on one or more microphone input signals. The controller is configured to control the processor based on the speech intelligibility indicator. The speech intelligibility estimator comprises a pitch estimator for estimating a pitch parameter of a first audio source. The speech intelligibility indicator is based on the pitch parameter and a direction of the first audio source.