scholarly journals Speech Enhancement for Hearing Impaired Based on Bandpass Filters and a Compound Deep Denoising Autoencoder

Symmetry ◽  
2021 ◽  
Vol 13 (8) ◽  
pp. 1310
Author(s):  
Raghad Yaseen Lazim AL-Taai ◽  
Xiaojun Wu
Keyword(s):  
Signal To Noise Ratio ◽  
Hearing Aid ◽  
Bandpass Filters ◽  
Fixed Number ◽  
Speech Quality ◽  
Hearing Impaired ◽  
Denoising Autoencoder ◽  
Perceptual Evaluation ◽  
Residual Noise ◽  
Two Stages

Deep neural networks have been applied for speech enhancements efficiently. However, for large variations of speech patterns and noisy environments, an individual neural network with a fixed number of hidden layers causes strong interference, which can lead to a slow learning process, poor generalisation in an unknown signal-to-noise ratio in new inputs, and some residual noise in the enhanced output. In this paper, we present a new approach for the hearing impaired based on combining two stages: (1) a set of bandpass filters that split up the signal into eight separate bands each performing a frequency analysis of the speech signal; (2) multiple deep denoising autoencoder networks, with each working for a small specific enhancement task and learning to handle a subset of the whole training set. To evaluate the performance of the approach, the hearing-aid speech perception index, the hearing aid sound quality index, and the perceptual evaluation of speech quality were used. Improvements in speech quality and intelligibility were evaluated using seven subjects of sensorineural hearing loss audiogram. We compared the performance of the proposed approach with individual denoising autoencoder networks with three and five hidden layers. The experimental results showed that the proposed approach yielded higher quality and was more intelligible compared with three and five layers.

scholarly journals Improving Speech Quality for Hearing Aid Applications Based on Wiener Filter and Composite of Deep Denoising Autoencoders

Signals ◽  
2020 ◽  
Vol 1 (2) ◽  
pp. 138-156
Author(s):  
Raghad Yaseen Lazim ◽  
Zhu Yun ◽  
Xiaojun Wu
Keyword(s):  
Hearing Loss ◽  
Signal To Noise Ratio ◽  
Hearing Aid ◽  
Contextual Information ◽  
Hybrid Approach ◽  
Wiener Filter ◽  
Speech Quality ◽  
Denoising Autoencoder ◽  
Noise Component ◽  
Perceptual Evaluation

In hearing aid devices, speech enhancement techniques are a critical component to enable users with hearing loss to attain improved speech quality under noisy conditions. Recently, the deep denoising autoencoder (DDAE) was adopted successfully for recovering the desired speech from noisy observations. However, a single DDAE cannot extract contextual information sufficiently due to the poor generalization in an unknown signal-to-noise ratio (SNR), the local minima, and the fact that the enhanced output shows some residual noise and some level of discontinuity. In this paper, we propose a hybrid approach for hearing aid applications based on two stages: (1) the Wiener filter, which attenuates the noise component and generates a clean speech signal; (2) a composite of three DDAEs with different window lengths, each of which is specialized for a specific enhancement task. Two typical high-frequency hearing loss audiograms were used to test the performance of the approach: Audiogram 1 = (0, 0, 0, 60, 80, 90) and Audiogram 2 = (0, 15, 30, 60, 80, 85). The hearing-aid speech perception index, the hearing-aid speech quality index, and the perceptual evaluation of speech quality were used to evaluate the performance. The experimental results show that the proposed method achieved significantly better results compared with the Wiener filter or a single deep denoising autoencoder alone.

scholarly journals Low-complexity artificial noise suppression methods for deep learning-based speech enhancement algorithms

2021 ◽  
Vol 2021 (1) ◽  
Author(s):  
Yuxuan Ke ◽  
Andong Li ◽  
Chengshi Zheng ◽  
Renhua Peng ◽  
Xiaodong Li
Keyword(s):  
Deep Learning ◽  
Speech Enhancement ◽  
Noise Suppression ◽  
Signal To Noise Ratio ◽  
Low Complexity ◽  
Speech Quality ◽  
Artificial Noise ◽  
Noise Power ◽  
Noise Masking ◽  
Residual Noise

AbstractDeep learning-based speech enhancement algorithms have shown their powerful ability in removing both stationary and non-stationary noise components from noisy speech observations. But they often introduce artificial residual noise, especially when the training target does not contain the phase information, e.g., ideal ratio mask, or the clean speech magnitude and its variations. It is well-known that once the power of the residual noise components exceeds the noise masking threshold of the human auditory system, the perceptual speech quality may degrade. One intuitive way is to further suppress the residual noise components by a postprocessing scheme. However, the highly non-stationary nature of this kind of residual noise makes the noise power spectral density (PSD) estimation a challenging problem. To solve this problem, the paper proposes three strategies to estimate the noise PSD frame by frame, and then the residual noise can be removed effectively by applying a gain function based on the decision-directed approach. The objective measurement results show that the proposed postfiltering strategies outperform the conventional postfilter in terms of segmental signal-to-noise ratio (SNR) as well as speech quality improvement. Moreover, the AB subjective listening test shows that the preference percentages of the proposed strategies are over 60%.

A Multiband Noise Reduction Wiener Filter Algorithm for Hearing Aid

2015 ◽  
Vol 719-720 ◽  
pp. 1082-1088
Author(s):  
Feng Jie Xue ◽  
Zhao Yang Guo ◽  
Xin An Wang
Keyword(s):  
Noise Reduction ◽  
Speech Processing ◽  
Speech Intelligibility ◽  
Hearing Aid ◽  
Wiener Filter ◽  
Voice Activity Detector ◽  
Perceptual Evaluation ◽  
Frequency Scale ◽  
Residual Noise ◽  
Digital Hearing Aid

Speech processing in hearing aid often operates in noisy environments. So noise reduction algorithm is very important for hearing aid. This paper reviews conventional wiener filter algorithm and points out its remained problems----musical residual noise and low speech intelligibility for low input signal-to-noise ratios (SNR). To solve the problems, the proposed algorithm divides the frequency (0-8000 Hz) into 16 bands in Mel-frequency scale and processes the signal separately to remove musical residual noise. Meanwhile, a new Voice Activity Detector (VAD) applied for digital hearing aid is also presented. Based on the accurate judgment of speech frame, the filter processes the speech frame and non-speech frame in different ways. The method is realized in matlab to test output SNR and PESQ (Perceptual Evaluation of Speech Quality). It is also realized in android app for real-time test. Both of the experimental results demonstrate the advances of the proposed algorithm.

scholarly journals Wearable Hearing Device Spectral Enhancement Driven by Non-Negative Sparse Coding-Based Residual Noise Reduction

Sensors ◽  
2020 ◽  
Vol 20 (20) ◽  
pp. 5751
Author(s):  
Seon Man Kim
Keyword(s):  
Speech Enhancement ◽  
Hearing Aids ◽  
Sparse Coding ◽  
Signal To Noise Ratio ◽  
A Priori ◽  
Wiener Filter ◽  
Unified Framework ◽  
Perceptual Evaluation ◽  
Residual Noise ◽  
Hearing Device

This paper proposes a novel technique to improve a spectral statistical filter for speech enhancement, to be applied in wearable hearing devices such as hearing aids. The proposed method is implemented considering a 32-channel uniform polyphase discrete Fourier transform filter bank, for which the overall algorithm processing delay is 8 ms in accordance with the hearing device requirements. The proposed speech enhancement technique, which exploits the concepts of both non-negative sparse coding (NNSC) and spectral statistical filtering, provides an online unified framework to overcome the problem of residual noise in spectral statistical filters under noisy environments. First, the spectral gain attenuator of the statistical Wiener filter is obtained using the a priori signal-to-noise ratio (SNR) estimated through a decision-directed approach. Next, the spectrum estimated using the Wiener spectral gain attenuator is decomposed by applying the NNSC technique to the target speech and residual noise components. These components are used to develop an NNSC-based Wiener spectral gain attenuator to achieve enhanced speech. The performance of the proposed NNSC–Wiener filter was evaluated through a perceptual evaluation of the speech quality scores under various noise conditions with SNRs ranging from -5 to 20 dB. The results indicated that the proposed NNSC–Wiener filter can outperform the conventional Wiener filter and NNSC-based speech enhancement methods at all SNRs.

scholarly journals Identificeerbaarheid van kinderen met een gehoorbeperking

2019 ◽  
Author(s):  
Nathalie Boonen ◽  
Hanne Kloots ◽  
Steven Gillis
Keyword(s):  
Cochlear Implant ◽  
Hearing Aid ◽  
Speech Quality ◽  
Hearing Impaired ◽  
Device Use ◽  
Hearing Children ◽  
Children's Speech ◽  
Impaired Children ◽  
Hearing Impaired Children

Abstract Studies on the speech and language development of hearing-impaired children often focus on (deviations in) the children’s speech production. However, it is unclear if listeners also perceive differences between the speech of normally hearing and hearing-impaired children. This contribution wants to fill this void by investigating the overall perceived speech quality of both groups. Three groups of listeners (speech and language pathologists, primary school teachers and inexperienced listeners) judged 126 utterances of seven normally hearing children, seven children with an acoustic hearing aid and seven children with a cochlear implant, in a comparative judgment task. All children were approximately seven years old and received, in the case of the hearing-impaired children, their assistive hearing device before the age of two. The online tool D-PAC was used to administer the comparative judgement task. The listeners compared stimuli in pairs and decided which stimulus sounded best. This method ultimately leads to a ranking in which all stimuli are represented according to their overall perceived speech quality. The main result is that the speech of normally hearing children was preferred by the listeners. This indicates that, even after several years of device use, the speech quality of hearing-impaired children is perceived as different from that of normally hearing children. Within the group of hearing-impaired children, cochlear implanted children were judged to exhibit higher speech quality than acoustically hearing aided children, especially after a longer device use. The speech quality of the latter group, on the other hand, remained practically stable. Listeners, irrespectively of their degree of experience with (hearing-impaired) children’s speech, completed the task similarly. In other words: the difference between the overall perceived speech quality of normally hearing and hearing-impaired children is salient for all listener groups and they all slightly preferred children with a cochlear implant over children with an acoustic hearing aid.

Hearing Aid Amplification Reduces Communication Effort of People With Hearing Impairment and Their Conversation Partners

2020 ◽  
Vol 63 (4) ◽  
pp. 1299-1311 ◽  
Author(s):  
Timothy Beechey ◽  
Jörg M. Buchholz ◽  
Gitte Keidser
Keyword(s):  
Hearing Impairment ◽  
Hearing Aid ◽  
Normal Hearing ◽  
Referential Communication ◽  
Hearing Impaired ◽  
Directional Hearing ◽  
Acoustic Environment ◽  
Communication Task ◽  
Acoustic Environments ◽  
The Impact

Objectives This study investigates the hypothesis that hearing aid amplification reduces effort within conversation for both hearing aid wearers and their communication partners. Levels of effort, in the form of speech production modifications, required to maintain successful spoken communication in a range of acoustic environments are compared to earlier reported results measured in unaided conversation conditions. Design Fifteen young adult normal-hearing participants and 15 older adult hearing-impaired participants were tested in pairs. Each pair consisted of one young normal-hearing participant and one older hearing-impaired participant. Hearing-impaired participants received directional hearing aid amplification, according to their audiogram, via a master hearing aid with gain provided according to the NAL-NL2 fitting formula. Pairs of participants were required to take part in naturalistic conversations through the use of a referential communication task. Each pair took part in five conversations, each of 5-min duration. During each conversation, participants were exposed to one of five different realistic acoustic environments presented through highly open headphones. The ordering of acoustic environments across experimental blocks was pseudorandomized. Resulting recordings of conversational speech were analyzed to determine the magnitude of speech modifications, in terms of vocal level and spectrum, produced by normal-hearing talkers as a function of both acoustic environment and the degree of high-frequency average hearing impairment of their conversation partner. Results The magnitude of spectral modifications of speech produced by normal-hearing talkers during conversations with aided hearing-impaired interlocutors was smaller than the speech modifications observed during conversations between the same pairs of participants in the absence of hearing aid amplification. Conclusions The provision of hearing aid amplification reduces the effort required to maintain communication in adverse conditions. This reduction in effort provides benefit to hearing-impaired individuals and also to the conversation partners of hearing-impaired individuals. By considering the impact of amplification on both sides of dyadic conversations, this approach contributes to an increased understanding of the likely impact of hearing impairment on everyday communication.

Singular Values Decomposition and Lifting Wavelet Transform for Speech Signal Embedding into Digital Image

2019 ◽  
Vol 12 (2) ◽  
pp. 138-151
Author(s):  
Mourad Talbi ◽  
Med Salim Bouhlel
Keyword(s):  
Wavelet Transform ◽  
Speech Signal ◽  
Signal To Noise Ratio ◽  
Perceptual Quality ◽  
Lifting Wavelet Transform ◽  
Signal To Noise ◽  
Perceptual Evaluation ◽  
Lifting Wavelet ◽  
Noise Ratio

Background: In this paper, we propose a secure image watermarking technique which is applied to grayscale and color images. It consists in applying the SVD (Singular Value Decomposition) in the Lifting Wavelet Transform domain for embedding a speech image (the watermark) into the host image. Methods: It also uses signature in the embedding and extraction steps. Its performance is justified by the computation of PSNR (Pick Signal to Noise Ratio), SSIM (Structural Similarity), SNR (Signal to Noise Ratio), SegSNR (Segmental SNR) and PESQ (Perceptual Evaluation Speech Quality). Results: The PSNR and SSIM are used for evaluating the perceptual quality of the watermarked image compared to the original image. The SNR, SegSNR and PESQ are used for evaluating the perceptual quality of the reconstructed or extracted speech signal compared to the original speech signal. Conclusion: The Results obtained from computation of PSNR, SSIM, SNR, SegSNR and PESQ show the performance of the proposed technique.

Suggestion of hearing aid gain for Korean Hearing Impaired (HAG-K)

2011 ◽  
Vol 7 (2) ◽  
pp. 119-123 ◽  
Author(s):  
Kyoung Won Lee ◽  
Jin Sook Kim
Keyword(s):  
Hearing Aid ◽  
Hearing Impaired

Magnitude Estimation of Loudness I Application to Hearing Aid Selection

1984 ◽  
Vol 27 (1) ◽  
pp. 20-27 ◽  
Author(s):  
Daniel Geller ◽  
Robert H. Margolis
Keyword(s):  
Magnitude Estimation ◽  
Hearing Aid ◽  
Normal Hearing ◽  
Hearing Impaired ◽  
Octave Band ◽  
Loudness Function ◽  
Comfortable Loudness ◽  
Tonal Stimuli

Three experiments were conducted to explore the utility of magnitude estimation of loudness for hearing aid selection. In Experiment 1 the loudness discomfort level (LDL), most comfortable loudness (MCL), and magnitude estimations (MEs) of loudness were obtained from normal-hearing subjects. MCLs fell within a range of loudnesses that was relatively low on the loudness function. The LDLs were lower than previously published values. Experiment 2 was performed to identify the source of disparity between our LDL data and previously reported results. The effects of instructions are demonstrated and discussed. In Experiment 3 magnitude estimations of loudness were used to determine the loudness of tonal stimuli selected to represent ⅓ octave band levels of speech. Over the 500–4000 Hz range, the contributions of the various frequency regions to the loudness of speech appears to be nearly constant. Methods are proposed for (a) predicting the frequency-gain response of a hearing aid that restores normal loudness for speech for the hearing-impaired listener and (b) psychophysically evaluating the compression characteristic of a hearing aid.

Sign in / Sign up

Export Citation Format

Share Document