Search

Summary This study aims to detect voice disorders related to vocal fold nodule, Reinke’s edema and neurological pathologies through multiband cepstral features of the sustained vowel /a/. Detection is performed between pairs of study groups and multiband analysis is accomplished using the wavelet transform. For each pair of groups, a parameters selection is carried out. Time series of the selected parameters are used as input for four classifiers with leave-one-out cross validation. Classification accuracies of 100% are achieved for all pairs including the control group, surpassing the state-of-art methods based on cepstral features, while accuracies higher than 88.50% are obtained for the pathological pairs. The results indicated that the method may be adequate to assist in the diagnosis of the voice disorders addressed. The results must be updated in the future with a larger population to ensure generalization.

Smartphones allow good quality recordings; however, it cannot be claimed that the acoustic parameters obtained from them are comparable to those obtained with a professional microphone. The objective of this study is to establish whether there are significant differences when comparing the values of six acoustic parameters obtained from recordings using four smartphones and a professional microphone. The Praat programme was used to obtain the acoustic parameters: f0, Jitter, Shimmer, HNR, Alpha Ratio and L1 - L0 of the recording of a sustained vowel /a/ using iPhone SE, iPhone 6, Samsung S8, Huawei Y7 and the Behringer ECM8000 microphone. The sample was made up of 26 men and 26 women, from 18 to 26 years old without declared voice pathology. The repeated sample ANOVA test was used to compare the values. All the equipment show reproducibility between consecutive repeated measurements. The parameters f0 and Jitter were the only ones that did not show significant differences between the smartphones and the professional microphone. None of the smartphones studied can replace the professional microphone in voice recording for the evaluation of the six parameters analysed, except for f0 and Jitter.

Automatic voice condition analysis systems have been developed to automatically discriminate pathological voices from healthy ones in the context of two disorders related to exudative lesions of Reinke’s space: nodules and Reinke’s edema. The systems are based on acoustic features, extracted from sustained vowel recordings. Reduced subsets of features have been obtained from a larger set by a feature selection algorithm based on Whale Optimization in combination with Support Vector Machine classification. Robustness of the proposed systems is assessed by adding noise of two different types (synthetic white noise and actual noise recorded in a clinical environment) to corrupt the speech signals. Two speech databases were used for this investigation: the Massachusetts Eye and Ear Infirmary (MEEI) database and a second one specifically collected in Hospital San Pedro de Alcantara (Caceres, Spain) for the scope of this work (UEX-Voice database). The results show that the prediction performance of the detection systems appreciably decrease when moving from MEEI to a database recorded in more realistic conditions. For both pathologies, the prediction performance declines under noisy conditions, being the effect of white noise more pronounced than the effect of noise recorded in the clinical environment.

SUMMARY Purpose Speech fundamental frequency (SFF) assessment is essential for all dysphonia patients to effectively evaluate the therapeutic effects of voice therapy, especially in patients with disturbances in their voice pitch due to mutational dysphonia, Reinke's edema, or as side effects of hormone therapy. A standard method of SFF measurement remains unknown. Speech tasks such as sustained vowel phonation, counting, reading passage, and spontaneous speech have generally been used for SFF measurements. Ideally, spontaneous speech best reflects SFF; however, this task has not yet been clearly defined and is limited with regard to its adaptation to a clinical setting. A reliable task for SFF measurement in Japanese, which corresponds to a speech task that most closely reflects the value that would be observed with typical spontaneous speech, has not been investigated. This study aimed to identify a reliable speech task by measuring the SFF values elicited by different widely used speech tasks in Japanese, and assess its reliability and coefficient of determination (R2). Methods Sixty healthy volunteers (30 men and 30 women; aged 19–30 years; mean age 22.5 years) were enrolled. All experimental procedures were performed in Japanese. The SFF values for the speech tasks were determined through the voice samples recorded using a Pulse Code Modulation (PCM) recorder. Each task, except spontaneous speech, was repeated five times, and the average fundamental frequency in each task was determined as the SFF. To assess the reliability of the SFF values across daily variations within individual speakers, the SFF measurements were repeated on two different days, separated by at least 1 week. Results The SFF values of sustained /a/ phonation, sustained vowel-average, counting, reading passage, and spontaneous speech had excellent reliability, in terms of their reproduction based on intraclass correlation. Significantly high SFF values were observed, in decreasing order, for sustained vowels-average, counting, reading passage, and spontaneous speech in both males and females. The highest R2 for spontaneous speech was that of reading passage in both males (R2 = 0.771) and females (R2 = 0.806) (P Conclusion When spontaneous speech was presented as a task most reflective of daily conversation, reading passage was determined to be the reliable task to assess the therapeutic effect of voice therapy in Japanese.

The Acoustic Voice Quality Index (AVQI) is an objective tool based on six acoustic parameters. It uses sustained vowel and continuous speech in the analyses and therefore it must be validated in different languages. In this study, the newest version of AVQI (03.01) with an extended continuous speech sample and improved internal consistency was validated to Finnish-speaking population. The study included 197 native Finnish-speaking voluntary participants, out of which 111 were patients from a phoniatric clinic and 86 were healthy voice users. A sustained vowel and a reading sample were recorded. Mean number of the syllables comparable to the 3 second sustained vowel was calculated from the reading samples. Sixteen voice specialists evaluated the overall voice quality of the voice samples with a four-point scale. Statistic analyses were performed to test the diagnostic accuracy between healthy and disordered voices in Finnish-speaking population. The number of syllables, comparable to 3 seconds of sustained vowel, was 31. The correlation between the AVQI scores and the overall voice quality was strong (Spearman’s rho 0.77, p= 0.01). The AVQI score 1.83 was the best to distinguish healthy and dysphonic voices. The study confirmed the AVQI03.01FIN version to be a good tool in voice disorder diagnostics in Finnish speaking population. publishedVersion

Summary Objectives This study aimed to estimate the intertext variability of smoothed cepstral peak prominence (CPPS), examine whether sound-processing techniques improved its variability and diagnostic capability, and evaluate the degree of intertext variability in detail with reference to the CPPS variabilities in sustained vowels. Study design This was a retrospective study. Methods Text readings of 58 Japanese syllables were recorded from 210 speakers with different diagnoses and varying degrees of dysphonia, and were divided into six passages. Applying the sound-processing techniques to those passages, we prepared three sample types: (1) nonprocessed, (2) only-loud, and (3) only-voiced samples. The intertext CPPS variability and diagnostic properties were compared across the passages and sample types. For detailed analysis, we subsequently extracted 63 normophonic speakers who maintained constant quality in their vowel utterances to evaluate the degree of intertext CPPS variability in relation to the variabilities between repeated identical vowels and across different vowels. Results Although several combinations of passages showed moderate-to-large CPPS variabilities, those variabilities were decreased by either technique, especially the deletion of silent segments, which resulted in the best diagnostic accuracy. The degree of intertext CPPS variability for the only-voiced samples was comparable to that of the CPPS variabilities in sustained vowels. Conclusions The sound-processing technique removing silent segments should be applied to enhance the diagnostic properties of CPPS. The additional technique of deleting unvoiced segments is worth adopting if clinicians and researchers seek to attenuate the influence of text differences in calculating CPPS values.

Objectives. Both acoustic and aerodynamic analyses are essential to evaluate the phonetic characteristics of voice pathology. The purpose of the study is to determine the magnitude of their correlation with the different types of bilabial plosive consonants.Methods. A controlled prospective study of 35 patients diagnosed with unilateral vocal fold paralysis was performed. The sustained vowel /a/ and bilabial voiceless consonants were used. Three common acoustic parameters were measured from a sustained vowel /a/ and aerodynamic parameters from a set of syllables /pi/, /phi/, and /p’i/. We determined the correlation coefficients between acoustic and aerodynamic measurements for the bilabial plosive consonants /pi/, /phi/, and /p’i/.Results. The mean values of acoustic parameters were higher than the thresholds of pathology. The mean values of aerodynamic parameters varied according to the types of consonants. The correlation between acoustic and aerodynamic parameters was significantly larger with the consonant /phi/ compared with the consonants /p’/ and /p/. The magnitudes of correlation were higher with the consonant /phi/ compared with the consonants /p’/ and /p/.Conclusion. The plosive consonant /phi/ may represent a more valuable investigative consonant than the consonants /p/ or /p’/ for aerodynamic analysis of voice pathology, especially in patients with unilateral vocal fold paralysis.

The paper presents an approach to the analysis of the modulation spectrum of a voice signal, in which the primary acoustic analysis is performed in bands of unequal width. Nonuniform analysis corresponds to the psychoacoustic laws of human perception of sound information. In the context of the analysis of the modulation spectrum, the considered approach can significantly reduce the resulting number of parameters, which greatly simplifies the task of detecting pathological changes in the voice signal based on the analysis of the parameters of the modulation spectrum. For frequency decomposition of a signal into bands of unequal width, two methods are considered: 1) DFT with channel combination and 2) the use of an nonuniform filter bank. The first method is characterized by a fixed time window for the analysis of all frequency components, while in the second method the time-frequency analysis plan is consistent with the critical frequency scale of the barks. For each method, a practical signal analysis circuit has been developed and described. The paper presents the experimental data on the application of the developed schemes for the analysis of the modulation spectrum to the problem of detecting pathology in a speech signal. The parameters of the modulation spectrum acted as information signs for a classifier built on the basis of linear discriminant analysis. Three different voice bases were used in the experiment (in two cases, the pathology was neurological ALS disease (amyotrophic lateral sclerosis), and in the third case, diseases of the larynx). The parameters of the modulation spectrum obtained in the DFT-based scheme with channel combining turned out to be more preferable for classification with a small number of features, however, greater accuracy (with an increase in the number of features) made it possible to obtain the parameters obtainedin the scheme based on an unequal filter bank. In all cases, the obtained classifiers were highly accurate (more than 97%). The obtained results show that the use of nonuniform time-frequency representation is preferable in the case when the analyzed signal is a sustained vowel phonation, since it provides higher accuracy of pathology detection using fewer modulation parameters

The objective of this study was to analyze the accuracy of recurrence quantification measurements (RQMs) in discriminating between individuals with and without voice disorders. This study consisted of a total of 541 recorded voice samples from normal and dysphonic subjects. All subjects recorded a sustained vowel /Ɛ/ and underwent a laryngoscopic examination of the larynx. Twelve RQMs and three parameters related to the topology of the phonatory system were extracted from the samples, for a total of 15 measures. The classification used quadratic discriminant analysis and includes the measures of accuracy, sensitivity, and specificity. Single measurements such as Shannon's entropy, average diagonal length, and transitivity had only acceptable performance ratings (≥70%) in discriminating between individuals with and without voice disorders. The combination of the parameters average diagonal length, Shannon's entropy, trapping time, length of the longest vertical line, tau, imbedding dimension, neighborhood radius, and transitivity produced the highest accuracy in discrimination (83.27%). Therefore, the performance of RQMs related to the formation of diagonal lines in classifying individuals with and without voice disorders was acceptable at ≥70%. A combination of RQMs showed good performance in discriminating between the study groups, with higher sensitivity and specificity.

Summary Objectives. The objective of this study was to investigate the ability of a two-stage method of cepstral peak identification to effectively discriminate rough vs breathy vs typical voice in sustained vowel productions. It was hypothesized that a dual-stage search for cepstral peak prominences (CPP's) above and below specified quefrency/F0 cutoffs would result in a CPP difference that would be characteristic of the rough, diplophonic voice type. Methodology. Central one-second portions of sustained vowel /a/ productions were obtained from 90 subjects (rough, breathy, and normophonic voices). All voice samples were analyzed using a a two-stage cepstral analysis process in which a CPP H i g h − L o w difference value was obtained by identifying cepstral peaks above and below a lower limit for expected F0 (150 Hz for females and 90 Hz for males), called CPPHigh and CPPLow respectively. Results. The CPP H i g h − L o w difference value was observed to be a highly significant predictor, with negative values for this parameter characteristic of a dominant subharmonic in the voice signal and the perception of diplophonic, rough voice. Correct classification of rough vs nonrough voice samples was 82.2% (sensitivity 0.80 and specificity 0.833). In the consideration of three group classification (breathy vs. normophonic vs. rough), models incorporating two predictors (the CPP obtained from a single search through a 60 to 300 Hz frequency range (CPPDefault) and the CPP H i g h − L o w difference value) correctly classified 78.88% of the voice samples. Conclusions. Rough, diplophonic voices were consistently observed to have a subharmonic peak that was greater in amplitude than the cepstral peak obtained within the region of the expected F0, resulting in a negative value for the CPP H i g h − L o w difference. The two-stage cepstral analysis process described herein is visually intuitive from the graphical display of a cepstrum and is a simple extended calculation derived from cepstral analysis procedures that have been recommended as essential in the acoustic description of vocal quality.

The authors are currently conducting research on methods to estimate psychiatric and neurological disorders from a voice by focusing on the features of speech. It is empirically known that numerous psychosomatic symptoms appear in voice biomarkers; in this study, we examined the effectiveness of distinguishing changes in the symptoms associated with novel coronavirus infection using speech features. Multiple speech features were extracted from the voice recordings, and, as a countermeasure against overfitting, we selected features using statistical analysis and feature selection methods utilizing pseudo data and built and verified machine learning algorithm models using LightGBM. Applying 5-fold cross-validation, and using three types of sustained vowel sounds of /Ah/, /Eh/, and /Uh/, we achieved a high performance (accuracy and AUC) of over 88% in distinguishing “asymptomatic or mild illness (symptoms)” and “moderate illness 1 (symptoms)”. Accordingly, the results suggest that the proposed index using voice (speech features) can likely be used in distinguishing the symptoms associated with novel coronavirus infection.

Summary Background Control of laryngeal muscles is required to manipulate pitch, volume, and voice quality. False vocal fold activity (FVFA) refers to the constriction and release of constriction of the false vocal folds. True vocal fold mass (TVFM) represents the cross-sectional thickness of the vocal folds. Larynx height (LH) refers to the vertical position of the larynx in the neck. To date, studies of voice control have examined the effects of these parameters separately. No study has investigated the impact of instructed systematic manipulation of these parameters on acoustic voice measures in vocally healthy trained subjects. Aims This study examined the effects of systematically manipulating FVFA, TVFM, and LH on several acoustic voice measures. Method Twelve vocally trained speakers were instructed to use specific techniques to achieve experimental conditions of constriction and release of constriction of FVFA, thicker and thinner TVFM, and normal and low LH. Each condition was implemented in combination with manipulating the other parameters. Voice recordings of sustained vowel /a/ and Rainbow Passage were obtained for all laryngeal manipulation conditions and underwent acoustic analyses for fundamental frequency (F0), signal typing, harmonics-to-noise ratio (HNR), cepstral peak prominence (CPP), and vocal relative intensity. Results Constricted FVFA caused more aperiodicity in the signals, lower CPP, and lower vocal relative intensity than release of constriction. Thicker TVFM resulted in significantly higher CPP and vocal relative intensity than thinner TVFM. Modifying TVFM did not affect F0 and HNR. Low LH had significantly lower F0 but did not impact on HNR, CPP, and intensity. Conclusions The effects of systematic manipulation of each laryngeal parameter resulted in independent acoustic effects without measurable interaction. Release of constriction of FVFA, thicker TVFM, and low LH were configurations that resulted in more optimal acoustic signals.

Acoustic analysis is widely used for assessing and monitoring vocal function. Dermatoglyphics is a method that analyzes genetic fingerprint markers, and uses that information for predicting physical skills related to anaerobic (explosive strength and speed) and aerobic (motor coordination and resistance) mechanisms. Therefore, it can be used as an indicator for individualized vocal training.To characterize the dermatoglyphic profiles and their relation to acoustic measures in voice professionals.A cross-sectional study involving 79 voice professionals. Acoustic analysis was performed using the PRAAT software. Mean intensity, fundamental frequency (F0), and cepstral peak prominence (CPP) value were extracted from the sustained vowel /a/. Fingerprints were collected using a Watson mini-integrated biometric scanner, and were quantified by design predominance, delta index (D10), total ridge count (TRC), and dermatoglyphic profile. The acoustic measures were analyzed descriptively and compared, considering the subjects' dermatoglyphic profiles. The confidence levels ranged from 90% to 95%.Most subjects exhibited anaerobic dermatoglyphic profiles (P = 0.004) and low TRC (p0.001). Higher F0 (P = 0.061), intensity (P = 0.065), and CPP (P = 0.073) were found for anaerobics (P0.001). There was a weak and negative correlation between TRC and intensity (P = 0.026), as well as between F0 (P = 0.017) and CPP (P = 0.069).Anaerobic profiles were predominant. The values of F0, intensity, and CPP increased for the anaerobics. There was a weak negative correlation between the TRC and intensity, F0, and CPP measures. Dermatoglyphics could have been seen as an interesting tool in the voice assessment and training direction for voice professionals.

RESUMO Objetivo Avaliar as modificações acústicas e de autopercepção obtidas após o primeiro, terceiro, quinto e sétimo minuto de prática da técnica de oscilação oral de alta frequência sonorizada, realizada com o dispositivo Shaker®. Método Participaram do estudo 27 mulheres com idade entre 18 e 41 anos com e sem queixa vocal. A vogal sustentada /ε/ foi registrada em tempo máximo de fonação antes (pré-exercício) e após o primeiro, terceiro, quinto e o sétimo minuto de execução da técnica de oscilação oral de alta frequência sonorizada com o dispositivo Shaker®. Foi realizada a análise acústica dos seguintes parâmetros: ruído, frequência fundamental, glottal to noise excitation, Jitter, Shimmer e número de harmônicos. Resultados Não houve diferenças significativas na comparação dos parâmetros acústicos entre as participantes com e sem sintomas vocais. Observou-se melhora na autopercepção do desconforto vocal nos grupos das mulheres com e sem sintomas vocais, comparando-se o momento antes da prática com o primeiro e terceiro minutos de prática. Houve redução dos valores de Jitter, ao comparar o momento antes da prática com os momentos após um e sete minutos e ao comparar os momentos após cinco e sete minutos de exercício, no grupo de mulheres com sintomas vocais. Conclusão A técnica com o Shaker® mostrou resultados positivos, tanto em indivíduos com sintomas quanto em indivíduos sem sintomas vocais. ABSTRACT Purpose To evaluate the acoustic and self-perception modifications obtained after the first, third, fifth and seventh minutes of voice oral high-frequency oscillation practice accomplished with the Shaker® device. Methodology Twenty-seven women aged between 18 and 41 years with and without vocal complaint participated in the study. The sustained vowel / ε / was recorded at maximum phonation time before (pre-exercise) and after the first, third, fifth and seventh minutes of voice oral high-frequency oscillation practice accomplished with Shaker®. The acoustic analysis of the following parameters was performed: noise, fundamental frequency, glottal to noise excitation, Jitter, Shimmer and number of harmonics. Results There were no significant differences in the comparison of the acoustic parameters between the participants with and without vocal symptoms. It was observed an improvement in the self-perception of vocal discomfort in the groups of women with and without vocal symptoms, comparing the moment before the practice with the first and third minutes of practice. There was a reduction in Jitter values when comparing the time before practice with the moments after one and seven minutes and when comparing the moments after five and seven minutes of exercise in the group of women with vocal symptoms. Conclusion The Shaker® technique showed positive results both in individuals with symptoms and in individuals without vocal symptoms.

Purpose: to assess the self-reported voice-related quality of life of schoolchildren without voice complaints and correlate it to acoustic parameters of voice. Methods: the research population comprised 31 children, mean age 6.5 (±0.17) years. The children’s perception of their voice-related quality of life was verified with the Pediatric Voice-Related Quality of Life Survey, which has 10 closed-ended questions and three domains. The acoustic parameters assessed were fundamental frequency, jitter, shimmer, glottal-to-noise excitation ratio, and noise rate, besides the phonatory deviation diagram, based on the analysis of the emission of the sustained vowel /ɛ/ for 5 seconds. Results: all the children obtained scores close to 100% in the three domains of the Pediatric Voice-Related Quality of Life Survey. As for the acoustic parameters, most of them presented abnormal values in the phonatory deviation diagram and in shimmer. There was a difference between girls and boys only in fundamental frequency. Conclusion: the pediatric self-reported voice-related quality of life of the children studied had a positive impact, despite the acoustic changes found in the voices. There was no correlation between the pediatric voice-related quality of life and the acoustic parameters in voice in the group studied. RESUMO Objetivo: avaliar a qualidade de vida em voz pediátrica autorreferida em escolares, sem queixas vocais e correlacioná-la aos parâmetros acústicos vocais. Métodos: a população desta pesquisa foi constituída por 31 crianças, com média de idade de 6,5 (±0,17) anos. Para verificar a percepção das crianças sobre a qualidade de vida em voz, foi aplicado o protocolo Qualidade de Vida em Voz Pediátrico composto por dez questões fechadas e três domínios. Os parâmetros acústicos avaliados foram frequência fundamental, jitter, shimmer, glottal-to-noise excitation ratio e índice de ruído, além do diagrama de desvio fonatório, a partir da análise da emissão da vogal /ɛ/ sustentada por cinco segundos. Resultados: os três domínios do protocolo Qualidade de Vida em Voz Pediátrico apresentaram escores próximos a 100% para todas as crianças. Quanto aos parâmetros acústicos, a maioria apresentou valores fora da normalidade, no diagrama de desvio fonatório e nos valores de shimmer. Observou-se diferença entre as meninas e os meninos apenas na frequência fundamental. Conclusão: a qualidade de vida em voz pediátrica autorreferida, nas crianças estudadas, apresentou impacto positivo, a despeito das alterações acústicas vocais encontradas. Não houve correlação entre a qualidade de vida em voz pediátrica e os parâmetros acústicos vocais, no grupo estudado.

Purpose There is a lack of standardized Mandarin reading text material that could potentially elicit significant variations in fundamental frequency (F0) and in vocal intensity for clinical voice evaluation. In this study, a phonetically balanced “Three Bears Passage” was developed based on the classical “Goldilocks” story for this purpose. The hypothesis was that the vocal range (F0 range and Intensity range) elicited while reading aloud the passage could be similar to that of the voice range profile (VRP) obtained with sustained vowel production and significantly different from that while reading an existing standard Mandarin passage. Method One hundred twenty young adults (60 men and 60 women, aged 25.22 ± 3.74 years) without vocal pathologies and vocal training participated in the study. Vocal range data, including F0-related measures and intensity-related measures, were obtained with Praat in order to construct VRPs for the sustained /a/ and to construct speech range profiles for passage reading. Results F0 range and intensity range for the new passage were significantly larger than those for the existing Mandarin passage for both male and female participants. In particular, the F0 range for the passage was not significantly different from that of the VRP for male participants. Conclusions These findings suggested that the new passage could allow individuals without vocal training to demonstrate a large vocal range in both F0 and vocal intensity, in a passage reading setting in a relatively short time. Further studies should be pursued with the passage to investigate the vocal range and vocal potential of individuals with voice disorders.

Objective To verify the modifications in auditory-perceptual and acoustic parameters of the fundamental frequency, short-term perturbation, noise, and cepstral measures of voice in young and elderly were distributed over the age of decades. Methods The study was conducted on 265 subjects (140 females and 125 males), aged 30 to 79 years, without a history of vocal complaints or dysphonia, distributed in five age groups for females and males (30-39 years until 70-79 years). Three speech therapists voice specialists performed the perceptual-auditory analysis using a 100-mm visual analog scale, based on the overall dysphonia grade (overall grade), roughness, and breathiness from the sustained vowel /a/. The smoothed cepstral peak prominence (CPPS), average fundamental frequency (F0), standard deviation of the fundamental frequency (std-F0), jitter percentage (%), shimmer percentage (%), noise-to-harmonic ratio (NHR), voice turbulence index (VTI), and soft phonation index (SPI) were extracted for the acoustic analysis Results In the case of the auditory-perceptual characteristics, the comparison between aged groups showed that, in men, the vocal quality did not show significant differences in the overall grade, roughness, and breathiness. However, women in the 3rd decade of life presented higher breathiness than those in the 5th, 6th, and 7th decades. Men had a higher overall grade and roughness than women. In the case of the acoustic data, the cepstral measure showed that in the 3rd decade, men presented higher CPPS than females. The noise measures, in the comparison by age groups, only VTI and SPI were different for women: VTI was higher in the 7th decade than in the 4th, whereas SPI was higher in the 4th decade than in older women. In the male population, the SPI in the 4th decade was also higher than that in the older population Conclusion The analysis by age groups of life allowed the identification of breathiness as a sensitive parameter in the different stages of adult life in women. The CPPS analysis showed that the 7th age decade is the period of the appearance of vocal changes for females, characterized by higher CPPS. The results related to traditional short-term perturbation and noise measurements showed that changes in F0, jitter, shimmer, and NHR were not found. Among the other noise measurements, VTI and SPI demonstrated some changes in different decades of life.

COVID-19, otherwise known as the coronavirus, has precipitated the world into a pandemic that has infected, as of the time of writing, more than 10 million persons worldwide and caused the death of more than 500,000 persons. Early symptoms of the virus include trouble breathing, fever and fatigue and over 60% of people experience a dry cough. Due to the devastating impact of COVID-19 and the tragic loss of lives, it is of the utmost urgency to develop methods for the early detection of the disease that may help limit its spread as well as aid in the development of targeted solutions. Coughs and other vocal sounds contain pulmonary health information that can be used for diagnostic purposes, and recent studies in chaotic dynamics have shown that nonlinear phenomena exist in vocal signals. The present work investigates the use of symbolic recurrence quantification measures with MFCC features for the automatic detection of COVID-19 in cough sounds of healthy and sick individuals. Our performance evaluation reveals that our symbolic dynamics measures capture the complex dynamics in the vocal sounds and are highly effective at discriminating sick and healthy coughs. We apply our method to sustained vowel ’ah’ recordings, and show that our model is robust for the detection of the disease in sustained vowel utterances as well. Furthermore, we introduce a robust novel method of informative undersampling using information rate to deal with the imbalance in our dataset, due to the unavailability of an equal number of sick and healthy recordings. The proposed model achieves a mean classification performance of 97% and 99%, and a mean \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$F_1$$\end{document}F1-score of 91% and 89% after optimization, for coughs and sustained vowels, respectively.

Many virological tests have been implemented during the Coronavirus Disease 2019 (COVID-19) pandemic for diagnostic purposes, but they appear unsuitable for screening purposes. Furthermore, current screening strategies are not accurate enough to effectively curb the spread of the disease. Therefore, the present study was conducted within a controlled clinical environment to determine eventual detectable variations in the voice of COVID-19 patients, recovered and healthy subjects, and also to determine whether machine learning-based voice assessment (MLVA) can accurately discriminate between them, thus potentially serving as a more effective mass-screening tool. Three different subpopulations were consecutively recruited: positive COVID-19 patients, recovered COVID-19 patients and healthy individuals as controls. Positive patients were recruited within 10 days from nasal swab positivity. Recovery from COVID-19 was established clinically, virologically and radiologically. Healthy individuals reported no COVID-19 symptoms and yielded negative results at serological testing. All study participants provided three trials for multiple vocal tasks (sustained vowel phonation, speech, cough). All recordings were initially divided into three different binary classifications with a feature selection, ranking and cross-validated RBF-SVM pipeline. This brough a mean accuracy of 90.24%, a mean sensitivity of 91.15%, a mean specificity of 89.13% and a mean AUC of 0.94 across all tasks and all comparisons, and outlined the sustained vowel as the most effective vocal task for COVID discrimination. Moreover, a three-way classification was carried out on an external test set comprised of 30 subjects, 10 per class, with a mean accuracy of 80% and an accuracy of 100% for the detection of positive subjects. Within this assessment, recovered individuals proved to be the most difficult class to identify, and all the misclassified subjects were declared positive; this might be related to mid and short-term vocal traces of COVID-19, even after the clinical resolution of the infection. In conclusion, MLVA may accurately discriminate between positive COVID-19 patients, recovered COVID-19 patients and healthy individuals. Further studies should test MLVA among larger populations and asymptomatic positive COVID-19 patients to validate this novel screening technology and test its potential application as a potentially more effective surveillance strategy for COVID-19.

Summary Objectives The present study firstly tries to find subgroups of pathological male and female phonation using data from a number of 534 pathological speakers. Secondly, this subgroup classification provides a basis for achieving voice profiles of pathological phonatory quality. Methods Using complementarily orientated electroglottographic and acoustic parametrization of phonatory quality, sustained vowel productions of 267 male and 267 female speakers were considered. Results In a first step, a clustering technique differentiates three subgroups within each gender on the basis of the EGG- and three subgroups on the basis of the acoustic parameters. In a second step, this subgroup definition allows one to present voice profiles of pathological speakers by combining the parameter means of the electroglottographically determined subgroups with those of the acoustically determined subgroups. Conclusions The presented voice profiles provide a finer reference basis for the classification of different pathological phonation types as well as for the evaluation of shifts in individual phonatory behavior due to therapy or spontaneous recovery.

Summary Objectives With the COVID-19 outbreak around the globe and its potential effect on infected patients’ voice, this study set out to evaluate and compare the acoustic parameters of voice between healthy and infected people in an objective manner. Methods Voice samples of 64 COVID-19 patients and 70 healthy Persian speakers who produced a sustained vowel /a/ were evaluated. Between-group comparisons of the data were performed using the two-way ANOVA and Wilcoxon's rank-sum test. Results The results revealed significant differences in CPP, HNR, H1H2, F0SD, jitter, shimmer, and MPT values between COVID-19 patients and the healthy participants. There were also significant differences between the male and female participants in all the acoustic parameters, except jitter, shimmer and MPT. No interaction was observed between gender and health status in any of the acoustic parameters. Conclusion The statistical analysis of the data revealed significant differences between the experimental and control groups in this study. Changes in the acoustic parameters of voice are caused by the insufficient airflow, and increased aperiodicity, irregularity, signal perturbation and level of noise, which are the consequences of pulmonary and laryngological involvements in patients with COVID-19.

OBJECTIVES. Report data on acoustic measures of voice in sustained vowels produced by typically developing children, aged 4 to 19 years, to add to the cross-sectional reference values in a pediatric database. METHODS. Recordings of sustained vowel /ɑ/ phonation were obtained from 158 children (80 males, 78 females) aged 4 to 19 years who were judged to be typically developing with respect to speech and voice. Acoustic analyses were performed with the Multidimensional Voice Program (MDVP(™)) and the Analysis of Dysphonia in Speech and Voice (ADSV(™)), both from Pentax Medical. RESULTS. Values from both MDVP and ADSV are reported for children in the following age cohorts: 4 to 6 years, 7 to 9 years, 10 to 12 years, 13 to 15 years, and 16 to 19 years. CONCLUSION. The data in this study complement previously published data and contribute to a pediatric reference database useful for research and for clinical practice related to children’s voice. Acoustic parameters most sensitive to age and sex are identified.

Purpose. The present study aimed at measuring the smoothed and non-smoothed cepstral peak prominence (CPPS and CPP) in teachers who considered themselves to have normal voice but some of them had laryngeal pathology. The changes of CPP, CPPS, sound pressure level (SPL) and perceptual ratings with different voice tasks were investigated and the influence of vocal pathology on these measures was studied. Method. Eighty-four Finnish female primary school teachers volunteered as participants. Laryngoscopically, 52.4% of these had laryngeal changes (39.3% mild, 13.1% disordered). Sound recordings were made for phonations of comfortable sustained vowel, comfortable speech, and speech produced at increased loudness level as used during teaching. CPP, CPPS and SPL values were extracted using Praat software for all three voice samples. Sound samples were also perceptually evaluated by five voice experts for overall voice quality (10 point scale from poor to excellent) and vocal firmness (10 point scale from breathy to pressed, with normal in the middle). Results. The CPP, CPPS and SPL values were significantly higher for vowels than for comfortable speech and for loud speech compared to comfortable speech (P 0.05). Conclusion. Neither the acoustic measures (CPP, CPPS, and SPL) nor the perceptual evaluations could clearly distinguish teachers with laryngeal changes from laryngeally healthy teachers. Considering no vocal complaints of the subjects, the data could be considered representative of teachers with functionally healthy voice.

Summary Objectives The primary objective was to analyze and compare the effects of the voiced high-frequency oscillation (VHFO) and Lax Vox techniques in different performance times. The secondary objective was to determine if there were any differences between the exercises in men and women. Methods Thirty volunteers (15 male, 15 female) without vocal complaints or any history of dysphonia participated in this study. The volunteers performed the VHFO and the Lax Vox techniques with a week washout, and the performance times were 1- and 3-minute long with a 15-minute interval. They answered two questionnaires on vocal and laryngopharyngeal symptoms intensity self-assessment and had acoustic measurements (cepstral peak prominence-smoothed – CPPs – , alpha ratio, and L1–L0) obtained, before 1 minute (M0), after 1 minute (M1), and after 3 minutes (M3). They had their sustained vowel /a/ and number counting 1–10 recorded. Data were analyzed by using the repeated measures ANOVA (P Results For vowel /a/, men had higher CPPs and lower alpha ratio values. For number counting, men had lower CPPs values, and for both genders there was an improvement after M1, which remained at M3. For the low-pitched voice symptom, men showed a decrease after M1, regardless of the exercise. Men had better results after VHFO, while women had better results after the Lax Vox technique for the tightness symptom. Regardless of exercise or gender, there was a decrease in tickling and irritated throat after M1, but at M3 their intensity tended to increase. The lump in the throat symptom decreased at M1 and remained at the same intensity at M3 regardless of exercise or gender. Conclusions It is important to be aware of the voice effects and the patient reports when requesting 3 minutes of both exercises. Thus, the conclusion is that the best performance time is 1 minute. For most outcomes, there was no difference between exercises and participants' genders. However, for the low-pitched voice symptom, men benefited more from both exercises, and for tightness, men benefited more from VHFO; in contrast, women benefited more from the Lax Vox technique. Therefore, such gender differences should be considered when choosing these exercises.

Introduction: Specific elements such as weight and body volume can interfere in voice production and consequently in its acoustic parameters, which is why it is important for the clinician to be aware of these relationships. Objective: To investigate the relationship between body mass index and the average acoustic voice parameters. Methods: Observational, cross-sectional descriptive study. The sample consisted of 84 women, aged between 18 and 40 years, an average of 26.83 (±6.88). The subjects were grouped according to body mass index: 19 underweight; 23 normal ranges, 20 overweight and 22 obese and evaluated the fundamental frequency of the sustained vowel [a] and the maximum phonation time of the vowels [a], [i], [u], using PRAAT software. The data were submitted to the Kruskal–Wallis test to verify if there were differences between the groups regarding the study variables. All variables showed statistically significant results and were subjected to non-parametric test Mann–Whitney. Results: Regarding to the average of the fundamental frequency, there was statistically significant difference between groups with underweight and overweight and obese; normal range and overweight and obese. The average maximum phonation time revealed statistically significant difference between underweight and obese individuals; normal range and obese; overweight and obese. Conclusion: Body mass index influenced the average fundamental frequency of overweight and obese individuals evaluated in this study. Obesity influenced in reducing maximum phonation time average. Resumo: Introdução: Elementos específicos como peso e volume corporal podem interferir na produção da voz e, consequentemente, em seus parâmetros acústicos, razão pela qual é importante que o médico esteja ciente dessas relações. Objetivo: Investigar a relação entre o índice de massa corporal e os parâmetros acústicos médios da voz. Método: Estudo observacional, transversal, descritivo. A amostra foi composta por 84 mulheres, com idade entre 18 e 40 anos, média de 26,83 (±6,88). As participantes foram agrupadas de acordo com o índice de massa corporal: 19 abaixo do peso; 23 com intervalos normais, 20 com sobrepeso e 22 obesos e avaliadas quanto a frequência fundamental da vogal sustentada [a] e o tempo máximo de fonação das vogais [a], [i], [u], utilizando o software PRAAT. Os dados obtidos foram submetidos ao teste de Kruskal-Wallis para verificar se houve diferença entre os grupos com relação as variáveis do estudo. Todas as variáveis apresentaram resultados estatisticamente significativos e foram submetidas ao teste não paramétrico de Mann-Whitney. Resultados: Com relação à média da frequência fundamental, houve diferença estatisticamente significativa entre os grupos com peso normal e sobrepeso e entre os grupos com peso normal e obesidade. A média do tempo máximo de fonação revelou diferença estatisticamente significativa entre indivíduos com baixo peso e obesidade; peso normal e obesidade; sobrepeso e obesidade. Conclusão: O índice de massa corporal influenciou na média da frequência fundamental dos indivíduos com sobrepeso e obesos avaliados neste estudo. A obesidade mórbida influiu na redução da média do tempo máximo de fonação. Keywords: Body mass index, Obesity, Phonation, Voice, Palavras-chave: Índice de Massa Corporal, Obesidade, Fonação, Voz

The acoustic analysis of speech has proved useful in the clinical evaluation of dysphonia, for it allows an objective assessment of the voice. However, the literature has suggested that the type of speech task used to obtain voice samples from patients (sustained vowel or connected speech) may affect both the perceptual and the acoustic evaluation of dysphonic voices. This study aimed at investigating whether the type of speech task significantly influences the acoustic analysis of dysphonic voices. Five acoustic parameters related to voice quality (cepstral peak prominence, difference between the magnitudes of the first and second harmonics, harmonics-to-noise ratio, jitter and shimmer) were automatically computed from voice samples of 5 female and 5 male subjects with and without dysphonia. These recordings consisted of three types of speech task: connected speech, count and sustained vowel. Analyses of variance with repeated measures showed that all five acoustic parameters were significantly affected by speech task. Further analyses through the Duncan’s multiple-range test indicated that the type of speech task may also influence the discrimination of dysphonic voices. It is concluded that speech task affects the acoustic assessment of dysphonic voices by significantly raising or reducing the values of the acoustic parameters.

Introduction and objectives In recent years, the use of cepstral measures for acoustic evaluation of voice has increased. One of the most investigated parameters is smoothed cepstral peak prominence (CPPs). The objectives of this paper are to establish the usefulness of this acoustic measure in the objective evaluation of alterations of the voice in Spanish and to determine what type of voice sample (sustained vowels or connected speech) is the most sensitive in evaluating the severity of dysphonia. Method Forty subjects participated in this study 40, 20 controls and 20 with dysphonia. Two voice samples were recorded for each subject (one sustained vowel/a/and four phonetically balanced sentences) and the CPPs was calculated using the Praat programme. Three raters perceptually evaluated the voice sample with the Grade parameter of GRABS scale. Results Significantly lower values were found in the dysphonic voices, both for/a/(t[38]=4.85, P Conclusion The results of this preliminary study suggest that CPPs is a good measure to detect dysphonia and to objectively assess the severity of alterations in the voice.

In this study, we wanted to discriminate between 30 patients who suffer from Parkinson’s disease (PD) and 20 patients with other neurological diseases (ND). All participants were asked to pronounce sustained vowel /a/ hold as long as possible at comfortable level. The analyses were done on these voice samples. Firstly, an initial feature vector extracted from time, frequency and cepstral domains. Then we used principal component analysis (PCA) and nonlinear PCA (NPCA). These techniques reduce the number of parameters and select the most effective ones to be used for classification. Support vector machine and k-nearest neighbor with different kernels was used for classification. We obtained accuracy up to 88% for discrimination between PD patients ND patients using KNN with k equal to three and five.

Fundamental frequency (F0) in speech is mainly defined by anatomical and biological factors, but it is also affected by different socio-cultural and behavioral models of voice usage. Unoptimal voice usage models can lead to voice disorders. This research examined the F0 of 16−17-year-old Finnish native speakers in reading and in spontaneous speech. In addition, the F0 in a sustained vowel as well as the reading speed was measured, and the students were asked to evaluate their own F0 on a VAS scale. The results were comparedwith Aino Sallinen-Kuparinen’s (1985) study, which stated that the speakers’ educational background and sex influence their F0.The speaking behaviour of Finnish students seems to have changed during the past 30 years: in the original study, girls attending a vocational school used a higher F0 than girls attending a high school. In addition, the speakers’ sex contributed to some differences in the students’ speaking behaviour. In the present study, neither the educational background nor the speaker’s sex had an effect on the pitch level. The girls’ F0 was 216 Hz in reading and 213 Hz in spontaneous speech. The boys had a low F0 compared to internationalstudies: 101 Hz in reading and 97 Hz in spontaneous speech. Possibly the boys in this age are affected by a social norm, which values a low speaking fundamental frequency. Low F0 was preferred also in the self-evaluation: the majority of the test persons estimated their own F0 to be lower than average.

Summary Objectives and Background Automatic voice pathology detection and classification systems effectively contribute to the assessment of voice disorders, which helps clinicians to detect the existence of any voice pathologies and the type of pathology from which patients suffer in the early stages. This work concentrates on developing an accurate and robust feature extraction for detecting and classifying voice pathologies by investigating different frequency bands using correlation functions. In this paper, we extracted maximum peak values and their corresponding lag values from each frame of a voiced signal by using correlation functions as features to detect and classify pathological samples. These features are investigated in different frequency bands to see the contribution of each band on the detection and classification processes. Material and Methods Various samples of sustained vowel /a/ of normal and pathological voices were extracted from three different databases: English, German, and Arabic. A support vector machine was used as a classifier. We also performed a t test to investigate the significant differences in mean of normal and pathological samples. Results The best achieved accuracies in both detection and classification were varied depending on the band, the correlation function, and the database. The most contributive bands in both detection and classification were between 1000 and 8000 Hz. In detection, the highest acquired accuracies when using cross-correlation were 99.809%, 90.979%, and 91.168% in the Massachusetts Eye and Ear Infirmary, Saarbruecken Voice Database, and Arabic Voice Pathology Database databases, respectively. However, in classification, the highest acquired accuracies when using cross-correlation were 99.255%, 98.941%, and 95.188% in the three databases, respectively.

The aim of the present study was to use electroglottography (EGG) to explore the effects of age and pitch level on sustained vowel phonation. Thirty female individuals (10 young, 10 middle-aged, and 10 older speakers) without voice disorders or training in singing participated in this study. Eight EGG parameters were measured during sustained vowel production with a high, mid, or low pitch: fundamental frequency, contact quotient, contacting-time quotient, decontacting-time quotient, speed quotient with a midslope criterion (SQ-mid), jitter, shimmer, and the harmonics-to-noise ratio. Age was found to be a significant factor in fundamental frequency, contact quotient, contacting-time quotient, decontacting-time quotient, and SQ-mid. With increasing age, the mean fundamental frequency decreased while the contact quotient increased. The middle-aged and older speakers had more asymmetrical vocal fold vibratory patterns than the young speakers. As for pitch level, the high pitch had a significantly less decontacting-time quotient and greater SQ-mid than low and mid pitches. The lack of significant interaction between age and pitch level indicates that the effects of age and pitch level could be additive. Finally, the discriminant analyses show that contact quotient is an important factor in predicting the age of a voice.

Acoustic Voice Quality Index (AVQI) has been reported to be a sensitive tool for discriminating between the normal and dysphonic voices, however, there is a dearth of research focused on investigating the diagnostic accuracy of AVQI in discriminating across the perceptual categories of dysphonia severity ie, slight, moderate, and severe. Hence, the present study is a preliminary attempt to document the AVQI across the degrees of dysphonia severity and to verify whether the AVQI can discriminate across these degrees of perceptual dysphonia severity.A total of 71 dysphonic samples extracted from the clinical database of the institute and 19 prospectively recorded normophonic samples consisting of sustained vowel /a/ and a sentence from a standardized Kannada passage served as input data. These 90 samples were categorized into four categories based on the overall grade of perceptual dysphonia (G) as normal, slight, moderate, or severe by three experienced speech language pathologists in a randomized and double-blinded manner, which served as the ground truth. The AVQI was calculated from the samples using Praat version 6.0.40 and the AVQI script version 2.03.The Spearman's correlation test revealed a positive correlation between the G and the AVQI (r = 0.67; P0.001). The Receiver Operating Characteristics analysis revealed that the AVQI could accurately discriminate across the perceptual subcategories of dysphonia.The results of the present study revealed that AVQI is a robust multiparametric measure which can accurately discriminate across the subcategories of the perceptual dysphonia severity with reasonable precision. Further, the AVQI is also shown to be promising in evaluating the signals with higher levels of aperiodicity such as severe hoarse voice quality.