scholarly journals Development Of Speech Material For An Armenian Speech Recognition Threshold Test

2021 ◽  
Vol 10 (3) ◽  
Author(s):  
Sona Sargsyan ◽  
Torsten Rahne
Keyword(s):  
Speech Recognition ◽  
Test List ◽  
Correlation Coefficients ◽  
Recognition Threshold ◽  
Speech Discrimination ◽  
Speech Corpus ◽  
Test Items ◽  
Threshold Test ◽  
Study Results ◽  
Homogeneous Test

Introduction — For the Armenian language, no standardized audiometric speech perception test is available. The purpose of this research was to develop, digitally record and evaluate an Armenian multisyllabic speech audiometry test, which can be used to measure speech recognition thresholds in Armenian native speakers. Material and Methods — To create a homogeneous multisyllabic speech corpus, Armenian numerals from 10-100 with 2-4 syllables were selected as general sample and digitally recorded by a female native Armenian speaker. For equalizing the speech recognition threshold between the test items, the speech discrimination function for each numeral was subsequently evaluated by five normal hearing native Armenian listeners in an experimental study. Results — Based on the phonemic structure of the Armenian language, 20 phonemically homogeneous test lists were created. The phoneme distributions of each test list correlated significantly and positively with that of the general sample (all Pearson moment correlation coefficients >0.960; all ps <0.001). Comparison of the phoneme distributions of test lists to that of the Armenian language showed that the test lists represent the language corpus quite well. After adjusting for actual threshold levels, speech discrimination functions are comparable between all numerals used. Conclusion — The developed test lists are a phonetically homogenous representation of the Armenian language and serve as an appropriated base for future clinical measurements of speech recognition threshold in Armenian speaking listeners.

scholarly journals Words in noise audiometry in adult subjects with normal hearing

2020 ◽  
Vol 50 (1) ◽  
pp. 9
Author(s):  
Widayat Alviandi ◽  
Jenny Bashiruddin ◽  
Brastho Bramantyo ◽  
Farisa Rizky
Keyword(s):  
Speech Recognition ◽  
Pure Tone ◽  
Pure Tone Audiometry ◽  
Normal Hearing ◽  
Recognition Threshold ◽  
Speech Audiometry ◽  
Discrimination Score ◽  
Speech Discrimination ◽  
Speech Discrimination Score ◽  
Noise Test

Background: Patients with hearing disturbance will generally undergo pure tone audiometry andspeech audiometry in a quiet room, but those examinations cannot evaluate the ability to understand speech in daily environment with a noisy background. Words in noise test will provide valuable informationregarding patient’s hearing problem in noise. Purpose: To evaluate the hearing threshold using wordsin noise test in adults with normal hearing. Method: This cross-sectional study was conducted in CiptoMangunkusumo Hospital from January to April 2017. All subjects who fulfilled the inclusion and exclusioncriteria underwent pure tone audiometry, speech audiometry, and words in noise test. Results: A total of71 individuals with normal hearing were recruited for this study. Words in noise test showed the medianvalue of 67 dB and 100 dB for Speech Recognition Threshold (SRT) 50% and Speech DiscriminationScore (SDS) 100%, respectively. The SRT 50% and SDS 100% were significantly higher in the age group40–60 years compared to the age group 18–39 years. There was also a statistically significant differencebetween males and females at SRT 50% assessed by words in noise audiometry. Conclusion: Wordsin noise test showed a statistically significant difference in SRT 50% and SDS 100% between two agegroups, but no difference was found between genders. The result of this study can be used as a referencefor SRT and SDS values of speech audiometry test in noise.Keywords: words in noise, speech audiometry, speech recognition threshold, speech discrimination score ABSTRAKLatar belakang: Pasien dengan gangguan pendengaran umumnya menjalani pemeriksaanaudiometri nada murni dan audiometri tutur di ruangan yang sunyi, tetapi pemeriksaan ini tidakdapat menggambarkan kemampuan pemahaman wicara di lingkungan sehari-hari yang ramai. Testutur dalam bising dapat mengevaluasi masalah pendengaran pasien dalam keadaan bising. Tujuan:Untuk mengevaluasi ambang pendengaran menggunakan tes tutur dalam bising pada orang dewasadengan pendengaran normal. Metode: Penelitian potong lintang ini dilakukan di Rumah Sakit CiptoMangunkusumo dari Januari hingga April 2017. Semua subjek yang memenuhi kriteria inklusi daneksklusi menjalani pemeriksaan audiometri nada murni, audiometri tutur, dan tes tutur dalam bising.Hasil: Sebanyak 71 orang dengan pendengaran normal diikutsertakan dalam penelitian ini. Tes tuturdalam bising menunjukkan nilai median masing-masing 67 dB dan 100 dB pada Speech RecognitionThreshold (SRT) 50% dan Speech Discrimination Score (SDS) 100%. SRT 50% dan SDS 100% secarasignifikan lebih tinggi pada kelompok usia 40–60 tahun dibandingkan dengan kelompok usia 18–39 tahun. Hasil pemeriksaan tes tutur dalam bising menunjukkan perbedaan yang signifikan antara laki-laki dan wanita pada nilai SRT 50%. Kesimpulan: Tes tutur dalam bising menunjukkan perbedaan yang bermakna secara statistik pada SRT 50% dan SDS 100% antara dua kelompok umur, tetapi tidak ada perbedaan signifikan diantara jenis kelamin. Hasil penelitian ini dapat digunakan sebagai acuan untuk nilai SRT dan SDS pada pemeriksaan audiometri tutur dalam bising.

Spanish Pediatric Speech Recognition Threshold Test

2019 ◽  
Vol 28 (3) ◽  
pp. 597-604 ◽  
Author(s):  
Lisa Lucks Mendel ◽  
Monique Pousson ◽  
Johnnie K. Bass ◽  
Rachel E. Lunsford ◽  
Caleb McNiece
Keyword(s):  
Speech Recognition ◽  
Pure Tone ◽  
Normative Data ◽  
Test Construction ◽  
Recognition Threshold ◽  
Stimulus Selection ◽  
Threshold Test ◽  
Spanish Speaking ◽  
Pure Tone Average ◽  
Study Test

Purpose The purpose of this study was to construct a recorded speech recognition threshold (SRT) test for Spanish-speaking children utilizing a picture board and a picture-pointing task. Design The Spanish Pediatric Speech Recognition Threshold (SPSRT) test was developed and validated in this study. Test construction steps included (a) stimulus selection, (b) assessment of familiarity, (c) digital recording, (d) creation of pictures that accurately depicted the target word from the stimulus set, and (e) validation of the test and recordings. SRTs were obtained from 24 Spanish-speaking children whose 1st language was Spanish. Results Normative data are presented that validate the SPSRT and establish the baseline relationship between the pure-tone average and the SRT obtained with the SPSRT. Results indicated that the SPSRT obtained using this test should be within 2–12 dB of an individual's pure-tone average for Spanish-speaking children with normal hearing and minimal hearing loss. Conclusions The SPSRT was developed and validated as a picture-pointing Spanish SRT test to be used with Spanish-speaking children. The 2-channel recording contains an English translation track, making this test easy to administer and interpret for clinicians without knowledge of Spanish.

Development of School Aged and Preschool Korean Bisyllabic Word Lists for Speech Recognition Threshold Test

2008 ◽  
Vol 4 (1) ◽  
pp. 37-47
Author(s):  
SooJin Cho ◽  
JungHak Lee ◽  
Dukhwan Lim ◽  
KyungWon Lee ◽  
HeeKyung Han
Keyword(s):  
Speech Recognition ◽  
Recognition Threshold ◽  
Threshold Test ◽  
Word Lists

Development of Korean Standard Bisyllabic Word List for Adults Used in Speech Recognition Threshold Test

2008 ◽  
Vol 4 (1) ◽  
pp. 28-36 ◽  
Author(s):  
Soo-Jin Cho ◽  
Dukhwan Lim ◽  
Kyung-Won Lee ◽  
Hee-Kyung Han ◽  
Jung-Hak Lee
Keyword(s):  
Speech Recognition ◽  
Word List ◽  
Recognition Threshold ◽  
Threshold Test

Growth performance, correlation and regression estimates of seven-chicken strains in South-Western Nigeria

2020 ◽  
Vol 45 (2) ◽  
pp. 40-48
Author(s):  
A. J. Atansuyi ◽  
U. C. Ihendu ◽  
C. A. Chineke
Keyword(s):  
Body Weight ◽  
Growth Performance ◽  
Correlation Coefficients ◽  
Tail Length ◽  
Experimental Period ◽  
The Body ◽  
Coefficient Of Determination ◽  
Indigenous Chicken ◽  
Randomized Design ◽  
Study Results

This study was conducted to determine the growth performance, correlation and regression estimates of seven-chicken strains in South-western Nigeria using a total of 300 day-old chicks. The birds were divided into seven groups based on their strain. The seven strains are Normal feather (NF), Fulani ecotype (FE), Frizzle feather (FF), naked neck (NN) and Transylvania indigenous strains while Hubbard and Marshal were meat-type exotic chickens. There were forty- five (45) unsexed day-old chicks in each strain except the Frizzle feather that were 30 in number. Completely randomized design (CRD) was used for the trial that lasted for 8 weeks. The birds were fed experimental diets ad libitum throughout the period of the study. Results showed that there were significant differences (p<0.05) in the initial and final weights of the birds. It was observed that exotic strains weighed heavier (3569.73gHB) than their indigenous counterparts (1391.11gNF). However, the Fulani ecotype weighed heaviest (1840.99g) among Nigeria indigenous strains during the experimental period. This showed that FE strains are generally heavy breed chicken and could be incorporated into a meat producing indigenous chicken if improved upon. The result of the correlation coefficients showed that a very strong, positive and highly significant (P<0.001) relationship existed between body weights and linear body measurements as most of the values are (>0.40). All the body parameter examined had significant (p<0.01) and direct relationship with the body weight. Shoulder-to-tail length (STL) had the highest coefficient of 0.98.The high correlation estimates obtained in this study could be as a result of pleiotropy, heterozygosity or linkage of genes in the birds. The three functions were highly significant (p<0.05) for all the parameters studied. This shows that the functions well described the parameters. On the basis of coefficient of determination (R ), the body weight of poultry birds at any age can be predicted most accurately with BRG using cubic function.

scholarly journals UCSY-SC1: A Myanmar speech corpus for automatic speech recognition

2019 ◽  
Vol 9 (4) ◽  
pp. 3194 ◽  
Author(s):  
Aye Nyein Mon ◽  
Win Pa Pa ◽  
Ye Kyaw Thu
Keyword(s):  
Neural Network ◽  
Speech Recognition ◽  
Automatic Speech Recognition ◽  
Gaussian Mixture ◽  
Error Rates ◽  
Training Data ◽  
Speech Corpus ◽  
Total Size ◽  
Test Sets ◽  
Web News

This paper introduces a speech corpus which is developed for Myanmar Automatic Speech Recognition (ASR) research. Automatic Speech Recognition (ASR) research has been conducted by the researchers around the world to improve their language technologies. Speech corpora are important in developing the ASR and the creation of the corpora is necessary especially for low-resourced languages. Myanmar language can be regarded as a low-resourced language because of lack of pre-created resources for speech processing research. In this work, a speech corpus named UCSY-SC1 (University of Computer Studies Yangon - Speech Corpus1) is created for Myanmar ASR research. The corpus consists of two types of domain: news and daily conversations. The total size of the speech corpus is over 42 hrs. There are 25 hrs of web news and 17 hrs of conversational recorded data.<br />The corpus was collected from 177 females and 84 males for the news data and 42 females and 4 males for conversational domain. This corpus was used as training data for developing Myanmar ASR. Three different types of acoustic models  such as Gaussian Mixture Model (GMM) - Hidden Markov Model (HMM), Deep Neural Network (DNN), and Convolutional Neural Network (CNN) models were built and compared their results. Experiments were conducted on different data  sizes and evaluation is done by two test sets: TestSet1, web news and TestSet2, recorded conversational data. It showed that the performance of Myanmar ASRs using this corpus gave satisfiable results on both test sets. The Myanmar ASR  using this corpus leading to word error rates of 15.61% on TestSet1 and 24.43% on TestSet2.<br /><br />

scholarly journals Factors Underlying Individual Differences in Speech-Recognition Threshold (SRT) in Noise Among Older Adults

2021 ◽  
Vol 13 ◽  
Author(s):  
Larry E. Humes
Keyword(s):  
Older Adults ◽  
Individual Differences ◽  
Speech Recognition ◽  
Detection Threshold ◽  
Recognition Threshold ◽  
Convenience Sample ◽  
Adult Intelligence Scale ◽  
Wechsler Adult Intelligence Scale ◽  
Group Data ◽  
Wechsler Adult Intelligence

Many older adults have difficulty understanding speech in noisy backgrounds. In this study, we examined peripheral auditory, higher-level auditory, and cognitive factors that may contribute to such difficulties. A convenience sample of 137 volunteer older adults, 90 women, and 47 men, ranging in age from 47 to 94 years (M = 69.2 and SD = 10.1 years) completed a large battery of tests. Auditory tests included measures of pure-tone threshold, clinical and psychophysical, as well as two measures of gap-detection threshold and four measures of temporal-order identification. The latter included two monaural and two dichotic listening conditions. In addition, cognition was assessed using the complete Wechsler Adult Intelligence Scale-3rd Edition (WAIS-III). Two monaural measures of speech-recognition threshold (SRT) in noise, the QuickSIN, and the WIN, were obtained from each ear at relatively high presentation levels of 93 or 103 dB SPL to minimize audibility concerns. Group data, both aggregate and by age decade, were evaluated initially to allow comparison to data in the literature. Next, following the application of principal-components factor analysis for data reduction, individual differences in speech-recognition-in-noise performance were examined using multiple-linear-regression analyses. Excellent fits were obtained, accounting for 60–77% of the total variance, with most accounted for by the audibility of the speech and noise stimuli and the severity of hearing loss with the balance primarily associated with cognitive function.

Sign in / Sign up

Export Citation Format

Share Document