Binaural Speech Intelligibility Predictor for Hearing Aid Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for evaluating the effectiveness of hearing aids are costly and time-consuming, relying on listening experiments that lack alternatives for assessing speech intelligibility in noisy environments and with hearing aid processing.
Innovation Solution
A binaural intrusive speech intelligibility measure is proposed, using four input signals (noisy and processed speech from both ears, and clean speech without noise or processing) to predict speech intelligibility, incorporating a hearing loss model and binaural advantage model to optimize speech intelligibility prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If listening experiments are used to evaluate hearing aid effectiveness, then measurement precision is improved, but loss of time and productivity deteriorate
Solution Approach 1:
The patent creates a computational model that copies the function of human listening experiments. The speech intelligibility predictor unit processes audio signals through algorithms that simulate how human listeners would evaluate speech intelligibility, providing the same measurement precision without requiring actual human subjects and lengthy experimental procedures
Solution Approach 2:
The patent replaces the mechanical system of human listening experiments with an automated computational system. The speech intelligibility predictor unit uses signal processing algorithms to automatically evaluate speech intelligibility in noisy environments, substituting human auditory evaluation with mechanical computation to eliminate time loss while maintaining measurement accuracy
2Measurement precision
If listening experiments are used to evaluate hearing aid effectiveness, then measurement precision is improved, but productivity deteriorates
Solution Approach 1:
The computational model copies the evaluative function of listening experiments, enabling rapid assessment of multiple hearing aid configurations and processing algorithms without the constraints of human subject availability and experimental scheduling, thereby dramatically improving evaluation productivity
Solution Approach 2:
The speech intelligibility predictor unit performs self-service evaluation by automatically processing audio signals and generating intelligibility metrics without requiring human intervention for each test case, allowing parallel processing of multiple evaluation scenarios and significantly boosting overall evaluation productivity
3Adaptability or versatility
If hearing loss models and binaural advantage models are incorporated, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent segments the speech intelligibility prediction task into distinct functional modules: a hearing loss model component that applies frequency-specific gains to simulate hearing impairment, and a binaural advantage model component that processes interaural time and level differences. This segmentation allows each model to be independently optimized and adjusted, improving adaptability to different listening conditions while managing device complexity through modular design
Data Source
Figure 1A~1D
Figure 2A
Figure 2B
AI summary
The application relates to an intrusive binaural speech intelligibility prediction system comprising a binaural speech intelligibility predictor unit adapted for receiving a target signal comprising speech in a) left and right essentially noise-free versions xl, xr, and in b) left and right noisy and/or processed versions yl, yr, said signals being received or being representative of acoustic signals as received at left and right ears of a listener. The binaural speech intelligibility predictor unit is configured to provide as an output a final binaural speech intelligibility predictor value SI measure indicative of the listener's perception of said noisy and/or processed versions yl, yr of the target signal. The application further relates to a method of providing a binaural speech intelligibility prediction value. The system comprises a) first, second, third and fourth input units for providing time-frequency representations xl(k,m), xr(k,m), yl(k,m) and yr(k,m) of said left and right noise-free versions and said left and right noisy and/or processed versions of the target signal, respectively, k being a frequency bin index, k=1, 2, ..., K, and m being a time index; b) a first Equalization-Cancellation stage adapted to receive and relatively time shift and amplitude adjust the left and right noise-free versions xl(k,m) and xr(k,m), respectively, and to provide a resulting noise-free signal x(k,m); c) a second Equalization-Cancellation stage adapted to receive and relatively time shift and amplitude adjust the left and right noisy and/or processed versions yl(k,m) and yr(k,m), respectively, and to provide a resulting noisy and/or processed signal y(k,m); and d) a monaural speech intelligibility predictor unit for providing final binaural speech intelligibility predictor value SI measure based on said resulting noise-free signal x(k,m) and said resulting noisy and/or processed signal y(k,m); wherein said first and second Equalization-Cancellation stages are adapted to optimize the final binaural speech intelligibility predictor value SI measure to indicate a maximum intelligibility of said noisy and/or processed versions yl, yr of the target signal by said listener. The invention may e.g. be used in development systems for hearing aids.