Simulations with FADE of the effect of impaired hearing on speech recognition performance cast doubt on the role of spectral resolution

Analysis Model of Spoken English Evaluation Algorithm Based on Intelligent Algorithm of Internet of Things

With the in-depth promotion of the national strategy for the integration of artificial intelligence technology and entity development, speech recognition processing technology, as an important medium of human-computer interaction, has received extensive attention and motivated research in industry and academia. However, the existing accurate speech recognition products are based on massive data platform, which has the problems of slow response and security risk, which makes it difficult for the existing speech recognition products to meet the application requirements for timely translation of speech with high response time and network security requirements under the condition of network instability and insecurity. Based on this, this paper studies the analysis model of oral English evaluation algorithm based on Internet of things intelligent algorithm in speech recognition technology. Firstly, based on the automatic machine learning and lightweight learning strategy, a lightweight technology of automatic speech recognition depth neural network adapted to the edge computing power is proposed. Secondly, the quantitative evaluation of Internet of things intelligent classification algorithm and big data analysis in this system is described. In the evaluation, the evaluation method of oral English characteristics is adopted. At the same time, the Internet of things intelligent classification algorithm and big data analysis strategy are used to evaluate the accuracy of oral English. Finally, the experimental results show that the oral English feature recognition system based on Internet of things intelligent classification algorithm and big data analysis has the advantages of good reliability, high intelligence, and strong ability to resist subjective factors, which proves the advantages of Internet of things intelligent classification algorithm and big data analysis in English feature recognition.

Modelling speech reception thresholds and their improvements due to spatial noise reduction algorithms in bimodal cochlear implant users

•

This paper compares two modelling approaches to predict the speech recognition ability of bimodal CI users and the benefit of using beamformers.

•

The modelling approaches vary in computational complexity and fitting requirements.

•

A complex cafeteria spatial scenario with three localized single noise source scenario and a diffuse multi-talker babble noise is used.

•

The automatic speech recognizer is more accurate across the different spatial scenarios and noise types and requires less fitting compared to the statistical modelling approach.

Spatial noise reduction algorithms (“beamformers”) can considerably improve speech reception thresholds (SRTs) for bimodal cochlear implant (CI) users. The goal of this study was to model SRTs and SRT-benefit due to beamformers for bimodal CI users. Two existing model approaches varying in computational complexity and binaural processing assumption were compared: (i) the framework of auditory discrimination experiments (FADE) and (ii) the binaural speech intelligibility model (BSIM), both with CI and aided hearing-impaired front-ends. The exact same acoustic scenarios, and open-access beamformers as in the comparison clinical study Zedan et al. (2021) were used to quantify goodness of prediction. FADE was capable of modeling SRTs ab-initio, i.e., no calibration of the model was necessary to achieve high correlations and low root-mean square errors (RMSE) to both, measured SRTs (r = 0.85, RMSE = 2.8 dB) and to measured SRT-benefits (r = 0.96). BSIM achieved somewhat poorer predictions to both, measured SRTs (r = 0.78, RMSE = 6.7 dB) and to measured SRT-benefits (r = 0.91) and needs to be calibrated for matching average SRTs in one condition. Greatest deviations in predictions of BSIM were observed in diffuse multi-talker babble noise, which were not found with FADE. SRT-benefit predictions of both models were similar to instrumental signal-to-noise ratio (iSNR) improvements due to the beamformers. This indicates that FADE is preferrable for modeling absolute SRTs. However, for prediction of SRT-benefit due to spatial noise reduction algorithms in bimodal CI users, the average iSNR is a much simpler approach with similar performance.

Thoughts on the potential to compensate a hearing loss in noise

Background: The effect of hearing impairment on speech perception was described by Plomp (1978) as a sum of a loss of class A, due to signal attenuation, and a loss of class D, due to signal distortion. While a loss of class A can be compensated by linear amplification, a loss of class D, which severely limits the benefit of hearing aids in noisy listening conditions, cannot. The hearing loss of class D is assumed to be the main reason why not few users of hearing aids keep complaining about the limited benefit of their devices in noisy environments. Working compensation strategies against it are unknown.

Methods: Recently, in an approach to model human speech recognition by means of a re-purposed automatic speech recognition (ASR) system, the loss of class D was explained by introducing a level uncertainty which reduces the individual accuracy of spectro-temporal signal levels. Based on this finding, an implementation of a patented dynamic range manipulation scheme (PLATT) is proposed which aims to mitigate the effect of increased level uncertainty on speech recognition in noise by expanding spectral modulation patterns in the range of 2 to 4 ERB. This compensation approach is objectively evaluated regarding the benefit in speech recognition thresholds in noise using the ASR-based speech recognition model. Recommendations for an evaluation with human listeners are derived.

Results: The objective evaluation suggests that approximately half of the class D loss due to an increased level uncertainty might be compensable. To measure the effect with human listeners, an experiment needs to be carefully designed to prevent the confusion class A and D loss compensations.

Conclusions: A working compensation strategy for the class D loss could provide previously unexploited potential for relief. Evidence has to be provided in experiments with human listeners.

Inference of the distortion component of hearing impairment from speech recognition by predicting the effect of the attenuation component

OBJECTIVE

A model-based determination of the average supra-threshold ("distortion") component of hearing impairment which limits the benefit of hearing aid amplification.

DESIGN

Published speech recognition thresholds (SRTs) were predicted with the framework for auditory discrimination experiments (FADE), which simulates recognition processes, the speech intelligibility index (SII), which exploits frequency-dependent signal-to-noise ratios (SNR), and a modified SII with a hearing-loss-dependent band importance function (PAV). Their attenuation-component-based prediction errors were interpreted as estimates of the distortion component.

STUDY SAMPLE

Unaided SRTs of 315 hearing-impaired ears measured with the German matrix sentence test in stationary noise.

RESULTS

Overall, the models showed root-mean-square errors (RMSEs) of 7 dB, but for steeply sloping hearing loss FADE and PAV were more accurate (RMSE = 9 dB) than the SII (RMSE = 23 dB). Prediction errors of FADE and PAV increased linearly with the average hearing loss. The consideration of the distortion component estimate significantly improved the accuracy of FADE's and PAV's predictions.

CONCLUSIONS

The supra-threshold distortion component-estimated by prediction errors of FADE and PAV-seems to increase with the average hearing loss. Accounting for a distortion component improves the model predictions and implies a need for effective compensation strategies for supra-threshold processing deficits with increasing audibility loss.

DARF: A data-reduced FADE version for simulations of speech recognition thresholds with real hearing aids

Developing and selecting hearing aids is a time consuming process which is simplified by using objective models. Previously, the framework for auditory discrimination experiments (FADE) accurately simulated benefits of hearing aid algorithms with root mean squared prediction errors below 3 dB. One FADE simulation requires several hours of (un)processed signals, which is obstructive when the signals have to be recorded. We propose and evaluate a data-reduced FADE version (DARF) which facilitates simulations with signals that cannot be processed digitally, but that can only be recorded in real-time. DARF simulates one speech recognition threshold (SRT) with about 30 min of recorded and processed signals of the (German) matrix sentence test. Benchmark experiments were carried out to compare DARF and standard FADE exhibiting small differences for stationary maskers (1 dB), but larger differences with strongly fluctuating maskers (5 dB). Hearing impairment and hearing aid algorithms seemed to reduce the differences. Hearing aid benefits were simulated in terms of speech recognition with three pairs of real hearing aids in silence (≥8 dB), in stationary and fluctuating maskers in co-located (stat. 2 dB; fluct. 6 dB), and spatially separated speech and noise signals (stat. ≥8 dB; fluct. 8 dB). The simulations were plausible in comparison to data from literature, but a comparison with empirical data is still open. DARF facilitates objective SRT simulations with real devices with unknown signal processing in real environments. Yet, a validation of DARF for devices with unknown signal processing is still pending since it was only tested with three similar devices. Nonetheless, DARF could be used for improving as well as for developing or model-based fitting of hearing aids.