Monaural Speech Intelligibility Predictor for Hearing Aids
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face challenges in predicting the intelligibility of noisy or processed speech signals without a noise-free reference signal, and in enhancing speech intelligibility in both monaural and binaural hearing scenarios.
Innovation Solution
A monaural speech intelligibility predictor unit processes time-frequency representations of speech signals to estimate intelligibility by normalizing and transforming time-frequency segments, using statistical methods and voice activity detection, and a binaural system combines monaural predictors to enhance intelligibility by adapting signal processing based on predicted intelligibility measures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If non-intrusive prediction method is used (without noise-free reference signal), then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent introduces an intermediary statistical model that represents the relationship between observed noisy time-frequency segments and the unobservable noise-free segments. This model acts as a mediator that allows prediction of intelligibility without direct access to noise-free reference signals, resolving the contradiction between ease of operation and measurement precision.
Solution Approach 2:
The patent creates a statistical copy or representation of the noise-free speech characteristics through the estimated parameters in the statistical model. Instead of requiring actual noise-free signals, the system uses this statistical copy to predict intelligibility, maintaining precision while improving operational ease.
2Device complexity
If statistical methods are used to estimate noise-free segments, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent transforms the complex problem of noise-free signal estimation into a parameter estimation problem. By changing the approach from direct signal processing to statistical parameter modeling, the system reduces computational complexity while maintaining measurement precision through the use of estimated parameters that capture essential speech characteristics.
3Device complexity
If monaural prediction is used, then device complexity is reduced, but measurement precision deteriorates compared to binaural
Solution Approach 1:
The patent creates a universal prediction framework that can operate in both monaural and binaural modes. The core statistical modeling approach remains the same, but the system can adapt to different input configurations (single channel or dual channel), providing multi-functionality that resolves the contradiction between device complexity and measurement precision.
Data Source
Figure 1A~1B
Figure 2A~2B
Figure 3A~3D
AI summary
Signal processing methods for predicting the intelligibility of speech, e.g., in the form of an index that correlate highly with the fraction of words that an average listener (amongst a group of listeners with similar hearing profiles) would be able to understand from some speech material are proposed. Specifically, solutions to the problem of predicting the intelligibility of speech signals, which are distorted, e.g., by noise or reverberation, and which might have been passed through some signal processing device, e.g., a hearing aid are described. In summary, the disclosure present solutions to the following problems: 1. Monaural, non-intrusive intelligibility prediction of noisy/processed speech signals 2. Binaural, non-intrusive intelligibility prediction of noisy/processed speech signals 3. Monaural and binaural intelligibility enhancement of noisy speech signals.