ABSTRACT
In this paper, we propose a multi-scale mel domain feature map extraction algorithm to solve the problem that the speech recognition rate of dysarthria is difficult to improve. We used the empirical mode decomposition method to decompose speech signals and extracted Fbank features and their first-order differences for each of the three effective components to construct a new feature map, which could capture details in the frequency domain. Secondly, due to the problems of effective feature loss and high computational complexity in the training process of single channel neural network, we proposed a speech recognition network model in this paper. Finally, training and decoding were performed on the public UA-Speech dataset. The experimental results showed that the accuracy of the speech recognition model of this method reached 92.77%. Therefore, the algorithm proposed in this paper can effectively improve the speech recognition rate of dysarthria.
Subject(s)
Humans , Dysarthria/diagnosis , Speech , Speech Perception , Algorithms , Neural Networks, ComputerABSTRACT
Digital and intelligent technology is an important driving force to promote the transformation and innovation of medical service mode and improve service quality. In February 2020, Peking Union Medical College Hospital established an intelligent outpatient pre-admission medical history collection system in order to improve the intelligent level of clinical application business system and the medical experience of outpatients.The system consisted of artificial intelligence capability layer, application service layer, capability resource layer and user application layer and hardware technology architecture. By docking with the hospital information system and using mobile terminal App, patients could submit pre-admission medical history information by themselves, and generate structured patient history for the visiting doctors to call. By November 2021, the pre-admission medical history collection system had covered 80% of the departments of the hospital, serving 1.68 million people in total, and the patient filling rate had reached 60%. The duration of patient′s medical history collection and the duration of doctor′s outpatient electronic medical record writing had been continuously shortened, greatly improving the efficiency of outpatient medical services.
ABSTRACT
Speech feature learning is the core and key of speech recognition method for mental illness. Deep feature learning can automatically extract speech features, but it is limited by the problem of small samples. Traditional feature extraction (original features) can avoid the impact of small samples, but it relies heavily on experience and is poorly adaptive. To solve this problem, this paper proposes a deep embedded hybrid feature sparse stack autoencoder manifold ensemble algorithm. Firstly, based on the prior knowledge, the psychotic speech features are extracted, and the original features are constructed. Secondly, the original features are embedded in the sparse stack autoencoder (deep network), and the output of the hidden layer is filtered to enhance the complementarity between the deep features and the original features. Third, the L1 regularization feature selection mechanism is designed to compress the dimensions of the mixed feature set composed of deep features and original features. Finally, a weighted local preserving projection algorithm and an ensemble learning mechanism are designed, and a manifold projection classifier ensemble model is constructed, which further improves the classification stability of feature fusion under small samples. In addition, this paper designs a medium-to-large-scale psychotic speech collection program for the first time, collects and constructs a large-scale Chinese psychotic speech database for the verification of psychotic speech recognition algorithms. The experimental results show that the main innovation of the algorithm is effective, and the classification accuracy is better than other representative algorithms, and the maximum improvement is 3.3%. In conclusion, this paper proposes a new method of psychotic speech recognition based on embedded mixed sparse stack autoencoder and manifold ensemble, which effectively improves the recognition rate of psychotic speech.
Subject(s)
Humans , Algorithms , Databases, Factual , Psychotic Disorders , Speech , Speech PerceptionABSTRACT
Abstract Introduction: Hearing acuity, central auditory processing and cognition contribute to the speech recognition difficulty experienced by older adults. Therefore, quantifying the contribution of these factors on speech recognition problem is important in order to formulate a holistic and effective rehabilitation. Objective: To examine the relative contributions of auditory functioning and cognition status to speech recognition in quiet and in noise. Methods: We measured speech recognition in quiet and in composite noise using the Malay Hearing in noise test on 72 native Malay speakers (60-82 years) older adults with normal to mild hearing loss. Auditory function included pure tone audiogram, gaps-in-noise, and dichotic digit tests. Cognitive function was assessed using the Malay Montreal cognitive assessment. Results: Linear regression analyses using backward elimination technique revealed that had the better ear four frequency average (0.5-4 kHz) (4FA), high frequency average and Malay Montreal cognitive assessment attributed to speech perception in quiet (total r2 = 0.499). On the other hand, high frequency average, Malay Montreal cognitive assessment and dichotic digit tests contributed significantly to speech recognition in noise (total r2 = 0.307). Whereas the better ear high frequency average primarily measured the speech recognition in quiet, the speech recognition in noise was mainly measured by cognitive function. Conclusions: These findings highlight the fact that besides hearing sensitivity, cognition plays an important role in speech recognition ability among older adults, especially in noisy environments. Therefore, in addition to hearing aids, rehabilitation, which trains cognition, may have a role in improving speech recognition in noise ability of older adults.
Resumo Introdução: A alteração da acuidade auditiva, do processamento auditivo central e da cognição são fatores que contribuem para a dificuldade de reconhecimento da fala em idosos. Portanto, quantificar a contribuição desses fatores no problema de reconhecimento da fala é importante para a formulação de uma reabilitação holística e efetiva. Objetivo: Examinar as contribuições relativas do funcionamento auditivo e do estado cognitivo para o reconhecimento da fala no silêncio e no ruído. Método: Nós medimos o reconhecimento de fala no silêncio e no ruído composto com o teste Malay hearing in noise test em 72 idosos malaios nativos falantes (60-82 anos) com audição normal a perda auditiva de grau leve. A avaliação da função auditiva incluiu audiograma de tons puros, teste gaps-in-noise e testes dicótico de dígitos. A função cognitiva foi avaliada pelo teste Malay Montreal cognitive assessment. Resultados: Análises de regressão linear com técnicas de eliminação backward na orelha melhor revelaram média de quatro frequências (0,5-4 kHz) (4AF), média de alta frequência e teste Malay Montreal cognitive assessment na orelha melhor, mensurada pela percepção da fala no silêncio (r2 total = 0,499). Por outro lado, a média de alta frequência, Malay Montreal cognitive assessment e o teste dicótico de dígitos contribuíram significativamente para o reconhecimento da fala no ruído (r2 total = 0,307). Enquanto a média de alta frequência da melhor orelha mediu principalmente o reconhecimento da fala no silêncio, o reconhecimento da fala no ruído foi mensurado principalmente pela função cognitiva. Conclusões: Esses achados destacam o fato de que, além da sensibilidade auditiva, a cognição desempenha um papel importante na capacidade de reconhecimento da fala em idosos, principalmente em ambientes ruidosos. Portanto, além de aparelhos auditivos, a reabilitação, que treina a cognição, pode ter um papel na melhoria da capacidade do reconhecimento da fala no ruído entre os idosos.
Subject(s)
Humans , Male , Female , Middle Aged , Aged , Presbycusis/physiopathology , Auditory Threshold/physiology , Speech Perception/physiology , Cognition/physiology , Noise , Presbycusis/diagnosis , Audiometry, Pure-Tone , Speech Reception Threshold Test , Acoustic Stimulation/methods , Hearing TestsABSTRACT
It is acknowledged that implantable bone conduction hearing aids such as a bone anchored hearing aid provide benefit and satisfaction to patients with unilateral hearing loss in terms of sound localization and speech perception under noise circumstances. This review paper introduces current models of the implantable bone conduction hearing aids, their candidacy, brief surgical procedures, and general audiological examinations. A table summary was used for nine published research papers which have very similar results by three points, i.e., no change in ability of sound localization, better performance of Hearing-In-Noise Test, and improved satisfaction of sound quality using questionnaires after the implantation. Unfortunately, systematic fitting strategy for users of the implantable bone conduction hearing aids has not been either studied or developed yet. Futhermore, there is no scientific based approach of reason why the implantation has no benefit in sound localization but still provides better performance for speech perception in noise. In conclusion, specific and effective fitting strategy of the current implantable bone conduction hearing aids should be developed and also support effect of the implantation with sensitivity and specificity for both clinicians and patients while accompanying their advanced technology.
Subject(s)
Humans , Bone Conduction , Hearing Aids , Hearing Loss, Unilateral , Noise , Sensitivity and Specificity , Sound Localization , Speech Perception , Suture AnchorsABSTRACT
Objective To investigate the speech recognition in noise in patients with unilateral conductive or mixed hearing impaired using Sophonotron Alpha 2 softband.Methods A total of 12 patients (12 ears) with unilateral conductive or mixed hearing loss participated in the survey and fitted with Sophonotron Alpha 2.The aided hearing threshold test and speech recognition test in noise were evaluated in both unaided and aided conditions.Results The average aided hearing threshold using Sophonotron Alpha 2 softband was 35.0±10.8 dB HL,showing significant improvement compared to that of unaided(P＜0.05).The average signal-to noise ratio loss was 5.9± 6.1 dB using Sophonotron Alpha 2 softband in noise,showing a more significant benefit than that of the unaided (13.6±10.9 dB).Conclusion The Sophonotron Alpha 2 softband can significantly improve speech recognition in noise for the patients with unilateral conductive or mixed hearing loss.
ABSTRACT
Objective The aim of this study is to evaluate word recognition abilities in 4-talker babble noise (BN) in children with normal hearing (NH), and to explore the lexical effects, age on speech recognition ability in children.Methods Children of 3~6 years old with normal hearing (N=212)were participated in this study.One hundred and twenty-four children were performed under the quiet conditions,while 88 were performed in 4-talker babble noise.The standard-Chinese version of the lexical neighborhood test was used to assess spoken word recognition performance.The speech level was 70 dB SPL, and signal to noise ratio (SNR) was chosen 4 dB in BN.Results The correct scores of the NH children in quiet for the DE, DH, ME, and MH words were 96.45%±5.17%, 88.87%±7.73%, 91.90%±7.31% and 82.38%±7.95%, while the scores in BN were 85.34%±11.23%, 66.42%±11.08%, 68.81%±15.99% and 48.58%±12.81%, respectively.The word-recognition performance in BN was significantly poorer than that in quiet.The word recognition scores showed that the lexical effects were significant in the BN (P<0.05).Children scored higher with dissyllabic words than with monosyllabic words;easy words scored higher than hard words in both quiet and noisy environments.The word-recognition performance also increased with age in each lexical category in the 3-to 6-year-old NH children.Conclusion Age and lexical characteristics of words had significant influences on the performance of Mandarin-Chinese word recognition in BN.The lexical effects were more noticeable under noise listening conditions than in quiet.The word-recognition performance in noise increased with age in NH children of 3-to 6-years old in noise and had not reached adult-like performance at 6 years of age.
ABSTRACT
Objective The purpose of this study is to investigate the effects of activating NLFC (nonlinear frequency compression, NLFC) on speech recognition in noise for listeners with high-frequency hearing loss.The results from the present study will provide guidance for clinical practice of fitting hearing aids to hearing impaired listeners.Methods Twenty-five hearing-impaired listeners participated in this study.They all wore binaural hearing aids.The subjects were divided into two groups based on whether NLFC was used in their hearing aids or not.Subjects who used NLFC were in group A and the others who used conventional process (CP) instead of NLFC were in group B.Both groups were tested with speech recognition in noise under the conditions of NLFC and CP.The statistical method of paired samples test was conducted to assess the effect of NLFC and CP on speech recognition in noise.Results The average performance in group A was 82.33%±16.06% and 76.70%±18.08 with NLFC and CP, respectively (P<0.01).It suggested that NLFC had a significant effect on subjects' speech recognition in noise in group A.However, the average performance in group B was 83.04%±12.56% and 81.79%±20.07% with NLFC and CP, respectively (P=0.19).It suggested that there was no significant effect of NLFC on subjects' speech recognition in noise in group B.The average high frequency threshold in group A was 53.54±7.30 and 57.01±6.81 dB SPL with NLFC and CP, respectively(P<0.01).And the average high frequency threshold in group B was 57.42±8.38 and 61.21±7.42 dB SPL with NLFC and CP, respectively(P=0.03).The significant correlativity in statistics on the speech recognition score in noise(y) verse the difference of high frequency hearing threshold (4,6,8 kHz) under the condition of NLFC activated and deactivated (x) in the subjects was y=1.21x-0.40 (r=0.63, t=3.89, P=0.007).Conclusion NLFC technology is beneficial to improving hearing aid users' audibility of high frequency and speech recognition in noise.Certain acclimatization period may be necessary to maximize the potential benefit for NLFC.
ABSTRACT
BACKGROUND AND OBJECTIVES: Listeners with sensorineural hearing loss want to compensate for their narrow dynamic range from appropriate compression functions of hearing aids. The present study aimed to determine which compression threshold and release time of nonlinear hearing aids might affect speech recognition and speech quality judgments. SUBJECTS AND METHODS: Ten elderly listeners with symmetrical moderate-to-severe sensorineural hearing loss participated in the study. To compare two compression threshold (31 dB SPL vs. 61 dB SPL) and two release time (50 ms vs. 500 ms) conditions, subjects’ speech recognition was measured using the Korean version of the Hearing in Noise Test in both quiet and noisy situations. All subjects were also asked to rate the degree of speech quality in terms of loudness, sharpness, clarity, and background noise immediately after completing the speech recognition test. RESULTS: Although no significant difference emerged in speech recognition thresholds between the two compression thresholds in the quiet situation, the compression threshold of 31 dB SPL showed a lower (or better) signal-to-noise ratio of speech recognition compared to that of 61 dB SPL. The release times of 50 and 500 ms did not statistically differ in both quiet and noisy situations. Subjective categories were found to be louder and clearer in the 50 ms release time for quiet and noisy, respectively. CONCLUSIONS: We expect that these patterns of current results will apply for better-fitting protocol of elderly Korean hearing aid users.
Subject(s)
Aged , Humans , Hearing Aids , Hearing Loss, Sensorineural , Hearing , Judgment , Noise , Signal-To-Noise RatioABSTRACT
BACKGROUND AND OBJECTIVES: Listeners with sensorineural hearing loss want to compensate for their narrow dynamic range from appropriate compression functions of hearing aids. The present study aimed to determine which compression threshold and release time of nonlinear hearing aids might affect speech recognition and speech quality judgments. SUBJECTS AND METHODS: Ten elderly listeners with symmetrical moderate-to-severe sensorineural hearing loss participated in the study. To compare two compression threshold (31 dB SPL vs. 61 dB SPL) and two release time (50 ms vs. 500 ms) conditions, subjects’ speech recognition was measured using the Korean version of the Hearing in Noise Test in both quiet and noisy situations. All subjects were also asked to rate the degree of speech quality in terms of loudness, sharpness, clarity, and background noise immediately after completing the speech recognition test. RESULTS: Although no significant difference emerged in speech recognition thresholds between the two compression thresholds in the quiet situation, the compression threshold of 31 dB SPL showed a lower (or better) signal-to-noise ratio of speech recognition compared to that of 61 dB SPL. The release times of 50 and 500 ms did not statistically differ in both quiet and noisy situations. Subjective categories were found to be louder and clearer in the 50 ms release time for quiet and noisy, respectively. CONCLUSIONS: We expect that these patterns of current results will apply for better-fitting protocol of elderly Korean hearing aid users.
Subject(s)
Aged , Humans , Hearing Aids , Hearing Loss, Sensorineural , Hearing , Judgment , Noise , Signal-To-Noise RatioABSTRACT
RESUMO Objetivo coletar normas de associação semântica de 96 palavras do português brasileiro e aplicá-las, posteriormente, na elaboração de um teste de reconhecimento de fala com frases com controle da previsibilidade da palavra. Método participaram 67 voluntários com idades entre 18 e 27 anos. Foi aplicada uma tarefa de associação semântica com delimitação de classe de palavras. Resultados o tamanho médio do conjunto total e do significativo foi maior para a segunda evocação para as classes de verbos e substantivos. A força de associação semântica que prevaleceu na primeira recordação foi forte para a primeira e média para a segunda palavra. Na segunda recordação, a força de associação com maior prevalência foi média para a primeira e para a segunda palavra. Foi observada correlação negativa significativa entre a força de associação e os tamanhos total e significativo do conjunto. Conclusão Normas de associação semântica de 96 palavras do português brasileiro foram coletadas conforme proposto e resultaram em palavras associadas semanticamente para cada alvo nas categorias de verbo e substantivo, as quais foram utilizadas em um teste de reconhecimento de fala com frases considerando a previsibilidade da palavra.
ABSTRACT Purpose To collect the semantic association norms of 96 Brazilian Portuguese words for further application in the formulation of a speech recognition test using sentences with controlled word predictability. Methods Study participants were 67 volunteers aged 18 to 27 years. A semantic association task with word class delimitation was used. Results The mean sizes of the total and meaning sets were larger in the second recall to both verb and noun classes. The prevalent semantic association strength in the first recall was strong to the first word and moderate to the second word. In the second recall, the prevalent semantic association strength was moderate to both the first and second words. Significant negative correlation was observed between association strength and total and meaning set sizes. Conclusion The semantic association norms of 96 Brazilian Portuguese words were collected as proposed and resulted in semantic associates for each target word in the verb and noun classes that were used in a speech recognition assessment considering sentences with controlled word predictability.
Subject(s)
Humans , Male , Female , Adult , Young Adult , Semantics , Speech Discrimination Tests/instrumentation , Vocabulary , Brazil , LanguageABSTRACT
Objective To evaluate Mandarin disyllables recognition scores in noise for normal hearing people , and to establish a model for teaohing .The second goal is to get the spatial separated advantage while the noise chan‐ging its direction .Methods Percentage of correct word recognition was measured for each list by testing 50 Manda‐rin-speaking people aged from 18 to 30 with normal aural/oral communicational abilities And .6 of them joined the pilot study aimed to identify a presentation level that would be used in the formal test .The other 44 subjects partici‐pated in the formal speech test .Results When the noise was at 0 and 90 ,the speech recognition changed along with the change of signal-to-noise ratio levels .Despite of the speech recognition effect ,there was a strong relation be‐tween the signal-to-noise ratio of 0° and 90° .Conclusion The direction of speech and noise may strongly influence the speech recognition scores .When the noise and signal is separated ,the score will be better .
ABSTRACT
Resumo:Este estudo tem como objetivo descrever, a partir de revisão de literatura, a confiabilidade da análise acústica utilizando o Multi Dimensional Voice Program,comparação de suas medidas com outros programas de análise acústica; caracterização de vozes de diferentes grupos; e sua utilização para verificar os efeitos e eficácia de diferentes procedimentos terapêuticos. Realizou-se levantamento bibliográfico que priorizaram estudos dos últimos cinco anos, incluindo-se artigos cujo objetivo estivesse de acordo com o interesse da presente revisão. A pesquisa foi realizada nos bancos de dados das bases Lilacs, BIREME, PubMed, MedLine, Scielo e Google Schoolar, por meio dos descritores acoustic, speech acoustics e voice. As medidas mais utilizadas são frequência fundamental, jitter, shimmer e proporção harmônico-ruído. As medidas de tremor não apresentam boa confiabilidade. As medidas mais consistentes que apresentam alta concordância com outros programas de análise acústica são as relacionadas com a frequência fundamental. Diversos estudos buscam caracterizar diferentes tipos de vozes tais como as de sujeitos sem alterações vocais de ambos os sexos, buscando estabelecer parâmetro de normalidade para diferentes grupos populacionais. Além disso, alguns estudos caracterizam vozes com diferentes distúrbios. A análise acústica é um recurso objetivo e necessário na avaliação de pacientes com distúrbio vocal e, principalmente, na comparação de diversos tipos de tratamento.
Abstract:The purpose of this study is to describe,the reliability of acoustic analysis using Multi Dimensional Voice Program, compare their measures with other acoustic analysis programs, characterize the voices of different groups, and use it to assess the effects and effectiveness of different therapeutic procedures. The study is based on a literature review that prioritized the past five years and included articles whose purpose was according to the interest of this review. The survey was conducted in the databases of Lilacs, BIREME, PubMed, Medline, SciELO and Google Scholar, using the descriptors acoustic, speech acoustics and voice. The measures most commonly used are fundamental frequency, jitter, shimmer and harmonics-to-noise ratio. The measures of tremor do not have much reliability. The most consistent measures that present a high correlation with other acoustic analysis programs are related to fundamental frequency. Several studies aim to characterize different types of voices such as those of subjects of both sexes without voice alterations, in order to establish a normality parameter for different population groups. In addition, some studies were characterized with different voice disorders. Acoustic analysis is an objective and necessary resource for the evaluation of patients with voice disorders and especially to compare several types of treatment.
ABSTRACT
Introducción: Las nuevas tecnologías son cada vez más aprovechadas por el sector de la salud en las intervenciones terapéuticas. Sin embargo, en el caso de los desórdenes comunicativos existen numerosas herramientas basadas en software libre que son poco conocidas por los terapeutas y que realmente podrían contribuir a la efectividad de sus intervenciones. Este trabajo sintetiza catorce herramientas de software libre que pueden apoyar las intervenciones fonoaudiológicas en estimulación y captura de atención, evaluación y manejo de la voz y el habla, recursos para comunicación aumentativa y alternativa y herramientas que facilitan la accesibilidad al computador. Materiales y métodos: La información aquí presentada es fruto de una revisión general de herramientas informáticas creadas para tratar los desórdenes de la comunicación humana. Se establecieron criterios de inclusión y de exclusión para seleccionar las herramientas y estas fueron instaladas y probadas. Resultados: De 22 herramientas encontradas se seleccionaron 14 las cuales fueron distribuidas en las categorías: Estimulación temprana y captura de atención, Procesamiento de la señal acústica de la voz, Procesamiento del habla, Comunicación Aumentativa y Alternativa y Otras; esta última incluye herramientas para actividades lúdicas y de acceso al computador sin la necesidad de conocimientos informáticos avanzados. Conclusión: En su conjunto, las herramientas tratadas en este trabajo proporcionan a terapeutas diversas alternativas libres para intervención, además, promueven la mejora de competencias informáticas tan necesarias en los profesionales de la sociedad actual.
Introduction: New technologies are increasingly used by the health sector for its implementation in therapeutic interventions. However, in the case of speech therapists, there are many unknown free software-based tools, which could support their daily work. This paper summarizes fourteen free software-based tools that can support interventions in early stimulation, assessment and control of voice and speech, several resources for augmentative and alternative communication and tools that facilitate access to the computer. Materials and methods: The information presented here is the result of a general review of software-based tools designed to treat human communication disorders. Criteria for inclusion and exclusion were established to select tools and these were installed and tested. Results: 22 tools were found and 14 were selected and classified in these categories: Early stimulation and attentional capture, acoustic signal voice processing, speech processing, Augmentative and Alternative Communication and Other; the latter includes tools for access to the computer without the need for advanced computer skills. Discussion: The set of tools discussed in this paper provides free computer-based tools to therapists in order to support their interventions. It additionally promotes the improvement of computer skills, so necessary in today's society of professionals.
Introdução. As novas tecnologias estão cada vez mais utilizado pelo setor de saúde para a sua implementação em intervenções terapêuticas. No entanto, no caso de terapeutas da fala, há muitas ferramentas baseadas em software livre desconhecidos que poderia apoiar o seu trabalho diário. Este artigo resume quatorze ferramentas baseadas em software livre que podem apoiar intervenções em Estimulação Precoce, avaliação e controle de voz e fala, vários recursos de comunicação e ferramentas aumentativa e alternativa que facilitem o acesso ao computador. Materiais e métodos: A informação aqui apresentada é o resultado de uma revisão geral de ferramentas baseadas em software projetados para tratar distúrbios da comunicação humana. Critérios de inclusão e exclusão foram estabelecidos para selecionar as ferramentas e estes foram instalados e testados. Resultados: 22 ferramentas foram encontradas e 14 foram selecionados e classificados nas seguintes categorias: A estimulação precoce e captura a atenção, processamento de sinal acústico de voz, processamento de voz, aumentativos e alternativos de comunicação e outros; este último inclui ferramentas para o acesso ao computador, sem a necessidade de conhecimentos avançados. Conclusão: O conjunto de ferramentas discutidas neste artigo fornece ferramentas baseadas em computador gratuitos para terapeutas, a fim de ajudar as suas intervenções, além disso, promove a melhoria de competências tão necessárias na sociedade de hoje de profissionais de informática.
Subject(s)
Humans , Software , Technology , Communication Disorders , Voice RecognitionABSTRACT
Objective To study normal hearing children's speech perception in Speech Spectrum -Shaped Noise (SSN) and Babble Noise (BN) using the Mandarin lexical neighborhood test .Methods Thirty -four children with normal hearing were included in this study and randomly assigned to the SSN group and the BN group .Each child was given the 12 lists of Mandarin lexical neighborhood test ,word recognition scores were acquired at different SNRs .Results SNR50 of dissyllablic easy word lists ,dissyllablic hard word lists ,monosyllable easy word lists , monosyllable hard word lists in SSN were -3 dB ,-0 .5 dB ,-1 dB and 3 .5 dB ,respectively ;SNR50 of the four categories lists in BN were -3 dB ,2 dB ,0 .5 dB and 10 dB ,respectively .Lexical effects had a significant influence on spoken word recognition in noise .Especially ,word recognition scores of easy words were higher than those of hard words ,dissyllablic words were better than monosyllable words .Conclusion The masking effects of babble noise on spoken speech perception are stronger than speech spectrum -shaped noise for 3-6 year-old normal hearing chil‐dren .Lexical effects also affet children's speech recognition in noise .
ABSTRACT
BACKGROUND AND OBJECTIVES: People usually converse in real-life background noise. They experience more difficulty understanding speech in noise than in a quiet environment. The present study investigated how speech recognition in real-life background noise is affected by the type of noise, signal-to-noise ratio (SNR), and age. SUBJECTS AND METHODS: Eighteen young adults and fifteen middle-aged adults with normal hearing participated in the present study. Three types of noise [subway noise, vacuum noise, and multi-talker babble (MTB)] were presented via a loudspeaker at three SNRs of 5 dB, 0 dB, and -5 dB. Speech recognition was analyzed using the word recognition score. RESULTS: 1) Speech recognition in subway noise was the greatest in comparison to vacuum noise and MTB, 2) at the SNR of -5 dB, speech recognition was greater in subway noise than vacuum noise and in vacuum noise than MTB while at the SNRs of 0 and 5 dB, it was greater in subway noise than both vacuum noise and MTB and there was no difference between vacuum noise and MTB, 3) speech recognition decreased as the SNR decreased, and 4) young adults showed better speech recognition performance in all types of noises at all SNRs than middle-aged adults. CONCLUSIONS: Speech recognition in real-life background noise was affected by the type of noise, SNR, and age. The results suggest that the frequency distribution, amplitude fluctuation, informational masking, and cognition may be important underlying factors determining speech recognition performance in noise.
Subject(s)
Adult , Humans , Young Adult , Cognition , Hearing , Masks , Noise , Railroads , Signal-To-Noise Ratio , VacuumABSTRACT
Objective To compare the results of manually -tested speech recognition threshold (SRT ) with automatically software -recorded SRT in the trial of Mandarin disyllabic test ,exploring the significance to the clini-cal applying .Methods 128 normal people of different ages without hearing loss and 57 workers exposed to noise in an automobile manufacturing was selected .These two group of volunteers speak mainly Mandarin in their daily life . MADSEN Conera (Danmark) clinical audiometr was applied .A group of double syllable word list with the same dif-ficulty of equivalence was used as test material .The initial presentation level was 20 dB HL higher than PTA .Then compared the results of manually -tested SRT with automatically software -recorded SRT .Results In the normal group ,the automatic value SRT was 7 .84 ± 3 .98 dB HL ,the manual value was 9 .19 ± 4 .47 dB HL ,and the average value of speech frequency threshold was 7 .63 ± 5 .78 dB HL .In the noise group ,the automatic value SRT was 6 .10 ± 8 .40 dB HL ,the manual value was 18 .81 ± 9 .52 dB HL ,and the average value of language frequency threshold was 27 .18 ± 19 .13 dB HL .There was significant difference between the values of SRT tested manually and recorded automatically (P<0 .01) .Conclusion There are differences between SRT valued manually and automatically .The SRT in people with normal hearing can be tested using automatic -recorded method .This method is convenient for screening in people without hearing loss .To exam in people with hearing loss ,the manual test is more appropriate .
ABSTRACT
BACKGROUND AND OBJECTIVES: The purpose of this study was to compare a newly developed Korean standard bisyllabic word list for adults (KS-BWL-A) against the conventional Hahm's List based on the reliability in speech recognition threshold (SRT) test. SUBJECTS AND METHOD: Twelve adults with normal hearing and 11 hearing-impaired patients participated in this study. After excluding 10 words that are common in both lists, 26 bisyllabic words from each list were used for comparison in this study. SRTs were obtained using the modified ascending method. RESULTS: The mean SRTs measured by KS-BWL-A were 1.9 dB lower for the normal hearing group and 2.7 dB lower for the hearing-impaired group than those measured by the conventional Hahm's list with statistical significance (p=0.025 and p=0.045). In both groups, the pure tone averages (PTAs) were highly correlated (correlation coefficient >0.7, p<0.001) with SRTs measured using old and new lists, and the differences between PTA and SRT were within 3 dB for both lists. CONCLUSION: These results suggest that both KS-BWL-A and the conventional list are reliable methods as adult SRT tests and that two lists are not considerably different. However, speech recognition was slightly easier by the new list than it was by the conventional list. Authors propose the KS-BWL-A as a standard word list for SRT measurement in Korea.
Subject(s)
Adult , Humans , Audiometry, Speech , Hearing , KoreaABSTRACT
OBJECTIVES: New signal processing technologies have recently become available for Baha(R) sound processors. These technologies have led to an increase in power and to the implementation of directional microphones. For any new technology, it is important to evaluate the degree of benefit under different listening situations. METHODS: Twenty wearers of the Baha osseointegrated hearing system participated in the investigation. The control sound processor was the Baha Intenso and the test sound processor was the Cochlear(TM) Baha(R) BP110power. Performance was evaluated in terms of free-field audibility with narrow band noise stimuli. Speech recognition of monosyllabic phonetically balanced (PB) words in quiet was performed at three intensity settings (50, 65, and 80 dB sound pressure level [SPL]) with materials presented at 0 degrees azimuth. Speech recognition of sentences in noise using the Hearing in Noise Test (HINT) in an adaptive framework was performed with speech from 0 degrees and noise held constant at 65 dB SPL from 180 degrees. Testing was performed in both the omni and directional microphone settings. Loudness growth was assessed in randomly presented 10 dB steps between 30 and 90 dB SPL to narrow band noise stimuli at 500 Hz and 3,000 Hz. RESULTS: The test sound processor had significantly improved high frequency audibility (3,000-8,000 Hz). Speech recognition of PB words in quiet at three different intensity levels (50, 65, and 80 dB SPL) indicated a significant difference in terms of level (P0.05). Speech recognition of sentences in noise demonstrated a 2.5 dB signal-to-noise ratio (SNR) improvement in performance for the test sound processor. The directional microphone provided an additional 2.3 dB SNR improvement in speech recognition (P<0.0001). Loudness growth functions demonstrated similar performance, indicating that both sound processors had sufficient headroom and amplification for the required hearing loss. CONCLUSION: The test sound processor demonstrated significant improvements in the most challenging listening situation (speech recognition in noise). The implementation of a directional microphone demonstrated a further potential improvement in hearing performance. Both the control and test sound processors demonstrated good performance in terms of audibility, word recognition in quiet and loudness growth.
Subject(s)
Humans , Bone Conduction , Hearing , Hearing Aids , Hearing Loss, Mixed Conductive-Sensorineural , Noise , Osseointegration , Sensory Aids , Signal-To-Noise RatioABSTRACT
OBJECTIVES: This study aimed to evaluate the benefits of ClearVoice strategy on speech perception in noise and in everyday listening situations in Cantonese-speaking cochlear implant users. METHODS: Twelve experienced adult users of the Harmony implant and HiRes 120 sound processing participated in the study. The study employed a prospective within-subjects design wherein speech recognition in adults using HiRes 120 without ClearVoice turned on (control option) was compared to their performance with HiRes 120 with ClearVoice turned on. Each subject was evaluated with two different ClearVoice gain settings: -12 dB (ClearVoice medium) and -18 dB (ClearVoice high) after one-week of use. The Cantonese hearing in noise test and a questionnaire were used as the outcome measures. RESULTS: Subjects performed significantly better with ClearVoice medium than with control option in noise. No significant difference in performance was noted among the 3 settings in quiet. Most subjects reported high level of satisfaction with ClearVoice in daily listening situations and preferred to keep ClearVoice on. CONCLUSION: ClearVoice can help cochlear implant recipients to hear better in noise.