scholarly journals A method for enhancing speech and warning signals based on parallel convolutional neural networks in a noisy environment

2021 ◽  
Vol 29 ◽  
pp. 141-152
Author(s):  
Ha Lim Kang ◽  
Sung Dae Na ◽  
Myoung Nam Kim
Keyword(s):  
Neural Networks ◽  
Hearing Loss ◽  
Hearing Aids ◽  
Noisy Environment ◽  
Warning Signals ◽  
Digital Hearing Aids ◽  
Sound Classification ◽  
Frequency Domains ◽  
Convolution Filter ◽  
Deep Learning Model

BACKGROUND: Digital hearing aids are based on technology that amplifies sound and removes noise according to the frequency of hearing loss in hearing loss patients. However, within the noise removed is a warning sound that alert the listener; the listener may be exposed to danger because the warning sound is not recognized. OBJECTIVE: In this paper, a deep learning model was used to improve these limits and propose a method to distinguish the warning sound in speech signals mixed with noise. In addition, the improved speech and warning sound were derived by removing noise present in the classification sound signals. METHODS: To classify the sound dataset, an adaptive convolution filter that changes according to two signals is proposed. The proposed convolution filter is applied to the PCNNs model to analyze the characteristics of the time and frequency domains of the dataset and classify the presence or absence of warning sound. In addition, the CEDN model was used to improve the intelligibility of the warning and the speech in the signal based on the warning sound classification from the proposed PCNNs model. RESULTS: Experimental results show that the PCNNs model using the proposed multiplicative filters is efficient for analyzing sound signals with complex frequencies. In addition, the CEDN model was used to improve the intelligibility of the warning and the speech in the signal based on the warning sound classification from the proposed PCNNs model. CONVLUSION: We confirmed that the PCNN model with the proposed filter showed the highest training rate, lowest error rate, and the most stable results. In addition, the CEDN model confirmed that speech and warning sounds were recognized, but it was confirmed that there was a limitation in clearly recognizing speech as the noise ratio increased.

Digital hearing aids for high-frequency sensorineural hearing loss: Preliminary experience with the RetroX® device

2005 ◽  
Vol 125 (7) ◽  
pp. 693-696 ◽  
Author(s):  
Maurizio Barbara ◽  
Giorgio Bandiera ◽  
Bruno Serra ◽  
Vania Marrone ◽  
Silvia Tarentini ◽  
...  
Keyword(s):  
Hearing Loss ◽  
Sensorineural Hearing Loss ◽  
Hearing Aids ◽  
High Frequency ◽  
Sensorineural Hearing ◽  
Digital Hearing Aids

Digital Hearing Aids Using Automated Varying Bandwidth Finite Impulse Response Filter Banks

2020 ◽  
Vol 10 (9) ◽  
pp. 2005-2009
Author(s):  
Hu Jun ◽  
Zeng Jian ◽  
Yi Hong ◽  
R. Jayabharathy ◽  
V. Gomathi ◽  
...  
Keyword(s):  
Hearing Loss ◽  
Auditory System ◽  
Hearing Aids ◽  
Filter Bank ◽  
Filter Banks ◽  
Finite Impulse Response ◽  
Research Work ◽  
Assistive Technologies ◽  
Digital Hearing Aids ◽  
Variable Bandwidth

A human auditory system is a highly complex sensitive system which transfers the acoustic sound into neuroelectrical signals toward the brain. Hearing difficulties or deafness are the outcomes of the problems occurred at any part of the auditory system. Assistive technologies such as hearing aids are developed to improve the quality of life of the hearing impaired people. Current digital hearing aids have fixed bandwidth filter banks which cannot provide enough flexibility to match with audiogram of different hearing loss. Recently, variable bandwidth filter banks have been introduced with different technologies to match more closely with the audiogram of a particular hearing loss. This research work proposes and implements a software controlled variable bandwidth FIR filter bank using Matlab GUI. In the Matlab GUI, in the filter bank, the bandwidth of every filter is adjusted dynamically in the Matlab GUI such that it fits more closely to the audiogram of a particular hearing loss. An experiment has been conducted in the developed Matlab GUI with various hearing loss, and the results show that the proposed system matches the filter bank magnitude response very closely to the audiogram of a particular hearing loss and reduces the matching error.

Hearing aids and tinnitus therapy: a 25-year experience

2008 ◽  
Vol 122 (10) ◽  
pp. 1052-1056 ◽  
Author(s):  
M I Trotter ◽  
I Donaldson
Keyword(s):  
Hearing Loss ◽  
Hearing Aids ◽  
Hearing Aid ◽  
University Teaching ◽  
Digital Hearing Aids ◽  
Subjective Tinnitus ◽  
Before And After ◽  
Digital Hearing Aid ◽  
Tinnitus Therapy ◽  
The Impact

AbstractObjectives:(1) To assess the subjective tinnitus perception of patients with audiologically proven hearing loss presenting to a tinnitus clinic, both before and after hearing aid provision; (2) to investigate subjective tinnitus perception in patients with unilateral and bilateral hearing loss; and (3) to assess the impact on tinnitus perception, if any, of a digital hearing aid programme in patients provided with hearing aids.Design:Prospective data collection for patients attending a tinnitus clinic over a 25-year period (1980–2004).Setting:University teaching hospital otolaryngology department.Participants:A total of 2153 consecutive patients attending a consultant-delivered specialist tinnitus clinic.Main outcomes measures:A visual analogue scale was used to assess the degree of tinnitus perception improvement, if any, comparing before versus after unilateral or bilateral aiding (in those with audiometrically proven hearing loss). A further assessment compared the effect of digital hearing aid programme introduction on symptomatic tinnitus perception in patients provided with unilateral or bilateral aids.Results:A total of 1440 patients were given hearing aids (826 unilateral and 614 bilateral). There was little difference in tinnitus perception, comparing overall aiding results in unilaterally or bilaterally aided patients. Overall, 554 (67 per cent) of unilaterally aided patients and 424 (69 per cent) of bilaterally aided patients reported some improvement in their tinnitus perception following aiding. There was a statistically significant improvement in tinnitus perception, comparing analogue aids with digital hearing aids, following introduction of a digital hearing aid programme in 2000, in both unilaterally (p < 0.001) and bilaterally (p < 0.001) aided patients.Conclusions:Provision of hearing aids in patients with audiometrically demonstrable hearing loss can play a very important part in tinnitus control. The additional improvement in tinnitus control observed following introduction of programmable digital aids had a summative effect in the management of these patients.

Hearing Aid Fitting Outcome: Clinical Application and Psychometric Properties of a Swedish Translation of the International Outcome Inventory for Hearing Aids (IOI-HA)

2010 ◽  
Vol 21 (08) ◽  
pp. 512-521 ◽  
Author(s):  
K. Jonas Brännström ◽  
Ingegerd Wennerström
Keyword(s):  
Hearing Loss ◽  
Psychometric Properties ◽  
Hearing Aids ◽  
Hearing Aid ◽  
Binaural Hearing ◽  
Demographic Variables ◽  
Sensorineural Hearing ◽  
Hearing Aid Fitting ◽  
Digital Hearing Aids ◽  
Conductive Hearing

Background: The International Outcome Inventory for Hearing Aids (IOI-HA) is a seven-item hearing-specific questionnaire. It was developed with the purpose of evaluating the efficacy of hearing aid rehabilitation. Few psychometric properties have been presented for a Swedish translation of the IOI-HA. Furthermore, previous studies have examined the IOI-HA in mainly sensorineural hearing losses, and we do not know how the type of hearing loss affects the outcome. Purpose: To evaluate the hearing aid fitting outcome measured in a clinical setting using a Swedish translation of the International Outcome Inventory for Hearing Aids (IOI-HA), to determine the psychometric properties of the translation, and to examine how a number of demographic variables such as type of hearing loss affect the outcome. Research Design: A descriptive and correlational study in a retrospective sample. Study Sample: Two hundred and twenty-four (107 females and 117 males; ages 27–94 yr with an average of 66.1 yr) first-time hearing aid users. Intervention: Mostly digital hearing aids (97.8%) were fitted monaurally (60%) or binaurally (40%) between 2007 and 2009. Data Collection and Analysis: The subjects were mailed the IOI-HA questionnaire six months after their final appointment, and the completed questionnaire was returned by mail to the clinic. The psychometric properties were evaluated and compared to previous studies using the IOI-HA. The associations between the outcome scores and a number of demographic variables (age, gender, degree of hearing loss, type of hearing loss, number of hearing aids, and type of hearing aids) were examined. Based on the pure tone audiograms, the subjects were divided into three groups; those with conductive hearing losses, sensorineural hearing losses, and mixed hearing losses. For these groups, the differences in outcome measured as IOI-HA were examined. Results: The psychometric properties of the present translation of the IOI-HA showed resemblance in many aspects to previous reports. Furthermore, the type of hearing loss seems to affect the IOI-HA outcome. Hearing loss increases with increasing age, and hearing aid use increases with increasing degree of hearing loss. Subjects with sensorineural hearing losses show significantly poorer scores on items concerning introspective aspects of the outcome in comparison to subjects with mixed hearing losses and subjects with conductive hearing losses. Monaurally fitted subjects tend to report lower scores on average, but monaural or binaural hearing aid fitting do not significantly affect the subjective outcome. Conclusions: The psychometric properties of the present Swedish translation of the IOI-HA show resemblance in many aspects to previous reports, but the differences observed could be due to differences in the study populations. Overall, the demographic variables examined could not be used as predictors for the hearing aid fitting outcome, and more reliable predictors need to be identified.

Evaluation of Nonlinear Frequency Compression for School-Age Children with Moderate to Moderately Severe Hearing Loss

2010 ◽  
Vol 21 (10) ◽  
pp. 618-628 ◽  
Author(s):  
Jace Wolfe ◽  
Andrew John ◽  
Erin Schafer ◽  
Myriel Nyffeler ◽  
Michael Boretzki ◽  
...  
Keyword(s):  
Hearing Loss ◽  
Speech Recognition ◽  
Hearing Aids ◽  
High Frequency ◽  
Nonlinear Frequency ◽  
Speech Sounds ◽  
Severe Hearing Loss ◽  
Frequency Compression ◽  
Digital Hearing Aids ◽  
Significant Difference

Background: Previous research has indicated that children with moderate hearing loss experience difficulty with recognition of high-frequency speech sounds, such as fricatives and affricates. Conventional behind-the-ear (BTE) amplification typically does not provide ample output in the high frequencies (4000 Hz and beyond) to ensure optimal audibility for these sounds. Purpose: To evaluate nonlinear frequency compression (NLFC) as a means to improve speech recognition for children with moderate to moderately severe hearing loss. Research Design: Within subject, crossover design with repeated measures across test conditions. Study Sample: Fifteen children, aged 5–13 yr, with moderate to moderately severe high-frequency sensorineural hearing loss were fitted with Phonak Nios, microsized, BTE hearing aids. These children were previous users of digital hearing aids and communicated via spoken language. Their speech and language abilities were age-appropriate. Data Collection and Analysis: Aided thresholds and speech recognition in quiet and in noise were assessed after 6 wk of use with NLFC and 6 wk of use without NLFC. Participants were randomly assigned to counter-balanced groups so that eight participants began the first 6 wk trial with NLFC enabled and the other seven participants started with NLFC disabled. Then, the provision of NLFC was switched for the second 6 wk trial. Speech recognition in quiet was assessed via word recognition assessments with the University of Western Ontario (UWO) Plural Test and recognition of vowel-consonant-vowel nonsense syllables with the Phonak Logatome test. Speech recognition in noise was assessed by evaluating the signal-to-noise ratio in dB for 50% correct performance on the Bamford-Kowal-Bench Speech-in-Noise (BKB-SIN) test, an adaptive test of speech perception in a multitalker babble background. Results: Aided thresholds for high-frequency stimuli were significantly better when NLFC was enabled, and use of NLFC resulted in significantly better speech recognition in quiet for the UWO Plural Test and for the phonemes /d/ and /s/ on the Phonak Logatome test. There was not a statistically significant difference in performance on the BKB-SIN test between the NLFC enabled and disabled conditions. Conclusions: These results indicate that NLFC improves audibility for and recognition of high-frequency speech sounds for children with moderate to moderately severe hearing loss in quiet listening situations.

scholarly journals Benefits from, Satisfaction with, and Self-Efficacy for Advanced Digital Hearing Aids in Users with Mild Sensorineural Hearing Loss

2018 ◽  
Vol 39 (02) ◽  
pp. 158-171 ◽  
Author(s):  
Anna Jilla ◽  
Jeffrey Danhauer ◽  
J. Sullivan ◽  
Kristin Sanchez ◽  
Carole Johnson
Keyword(s):  
Hearing Loss ◽  
Sensorineural Hearing Loss ◽  
Health Care Providers ◽  
Hearing Aids ◽  
Self Efficacy ◽  
Background Noise ◽  
Hearing Aid ◽  
Sensorineural Hearing ◽  
Care Providers ◽  
Digital Hearing Aids

AbstractLittle evidence is available regarding outcomes of advanced digital technology (ADT) hearing aid wearers with mild sensorineural hearing loss (MSNHL). The purpose of this article is to report the characteristics of and outcomes for this population. A cross-sectional research design was employed with 56 participants from a private practice setting. The International Outcomes Inventory for Hearing Aids (IOI-HA), Satisfaction with Amplification in Daily Life (SADL), and the Measure of Audiologic Rehabilitation Self-Efficacy for Hearing Aids (MARS-HA) were completed, scored, and compared with normative data. Results revealed that ADT aids were worn 10.5 hours/day, were mostly advanced to premium (55%), had an average cost per aid of $2,138 (SD = $840), and provided significant benefit (IOI-HA overall score: mean = 4.1; SD = 0.6) and satisfaction (SADL global score: mean = 5.4; SD = 0.8) to users who had good overall self-efficacy (MARS-HA composite score: mean = 81.7; SD = 12.8). Patients were most dissatisfied with and had the least self-efficacy for managing background noise and advanced handling of their devices. ADT hearing aid users with MSNHL achieved excellent outcomes, but ongoing follow-up and counseling from hearing health care providers may be important for successful management of background noise and mastery of advanced handling skills.

Moving to Mainstream Preschool

2011 ◽  
Vol 21 (2) ◽  
pp. 41-49
Author(s):  
Becky Clem
Keyword(s):  
Hearing Loss ◽  
Hearing Aids ◽  
Sound Field ◽  
Neural Pathways ◽  
Clear Speech ◽  
Digital Hearing Aids ◽  
Conversational Skills ◽  
The Face ◽  
Increased Sensitivity ◽  
Children With Hearing Loss

Children with hearing loss have a world of opportunities available to them because of marvelous advances in hearing technology, including super-powered digital hearing aids, cochlear implants, wireless FM systems, infrared classroom technology for sound field, and the bone-anchored hearing aid (BAHA). These advances, along with the ability to identify children with hearing loss soon after birth and fit them quickly with hearing technology, have changed the face of intervention and education. It is important for clinicians to start listening and spoken language (LSL) intervention very early, because babies' brains' neural pathways have increased sensitivity to sound before age 3 years. When children with hearing loss learn to speak through listening, they may stay on a normal path to speech and language development (Cole & Flexer, 2007). Now, children with hearing loss who learn to listen and speak typically develop regional accents, clear speech production, the ability to grasp idioms and jokes, and competent conversational skills. The purpose of this article is to describe the processes that are typical as a child with hearing loss transitions from services provided by professionals at an outpatient cochlear implant center to services provided in school settings.

Sign in / Sign up

Export Citation Format

Share Document