Pathological Voices Research Articles

Uncertainty handling using Improvised Intuitionistic fuzzy ANN based Voice Disorder Detection

The voice pathology detection is one of the essential process which has to be determined in the earlier stages because it is a sign for raising health related problems. The aim of this paper is to handle the uncertainty in voice dataset due to inconsistency in extracting potential features and vagueness in dealing voice signals. The raw voice signals are preprocessed by feature extraction using meyer wavelet and potential features involved in voice disorder detection are done using sequential forward feature selection methods as voice preprocessing. This research work introduced an improvised intuitionistic fuzzy artificial neural network which enhances the process of voice disorder detection is SVD database by using analytical hierarchical processing for assigning weights and thus the complete neural network performance was fine tuned instead of assigning the weights randomly. The simulation results proved the performance of the proposed model as best by producing more promising result while comparing with ANN, PANN and Fuzzy ANN models.

Analysis of Fundamental Frequency, Jitter and Shimmer in Stuttered and Non-Stuttered Speech of Marathi Language

Stuttering is one of the complex speech disorder which lead speech unintelligible to other and ASR system. In this paper Fundamental frequency (f0), vocal jitter and shimmer were measures the numerical words of 17 Stuttering (PWS) (i.e. 13 Male and 4 female) and 25 Non-Stutterers (NS) (i.e. 15 male 10 female) belongs to the Marathi Language. Jitter and shimmer are measurements of the cycle - to - cycle variations of F0 frequency and amplitude, respectively, which were largely used to describe pathological voice quality. The experiment performed with Pasrelmouth Praat Scripts in Python. This is the first attempt for the Marathi Language. Shimmer were shows the negligible difference between PWS and NS speech samples.

Applicability of Resonance Tube Phonation in Water for Semi-Occluded Vocal Tract Voice Training and Therapy

Background: Exercises involving tube phonation in water (TPW) have emerged as an alternative to conventional voice therapy techniques. The objective was to determine whether the results of these techniques were comparable to each other for the treatment of voice pathology, as well as to compare the costs implicated in the delivery of each treatment. Methods: A retrospective, descriptive, and observational study was performed that included all patients evaluated in our voice unit in 2015 who received vocal rehabilitation treatments. They were divided into two groups depending on the treatment technique used. The patients’ voice was evaluated subjectively using the GRBAS scale. We compared the clinical results and the specific costs of each technique. Results: A total of 55 patients were studied, of which 28 followed TPW exercises, and 27 underwent supervised vocal rehabilitation (SVR). Most of the patients who received TPW therapy were diagnosed with functional dysphonia, whereas functional-organic dysphonia was the most common pathology in the SVR group. Significant differences were evident when the total GRBAS scores were compared before and after the treatments, yet not when the GRBAS scores were compared between the two rehabilitation techniques. The incremental cost-effectiveness analysis revealed that TPW was less expensive than SVR, with both techniques proving to be equally effective in improving the pathological problems. Conclusions: TPW is similarly effective as SVR techniques in treating voice pathologies, yet it is significantly cheaper in terms of healthcare costs.

Impact of Subharmonic and Aperiodic Laryngeal Dynamics on the Phonatory Process Analyzed in Ex Vivo Rabbit Models.

Normal voice is characterized by periodic oscillations of the vocal folds. On the other hand, disordered voice dynamics (e.g., subharmonic and aperiodic oscillations) are often associated with voice pathologies and dysphonia. Unfortunately, not all investigations may be conducted on human subjects; hence animal laryngeal studies have been performed for many years to better understand human phonation. The rabbit larynx has been shown to be a potential model of the human larynx. Despite this fact, only a few studies regarding the phonatory parameters of rabbit larynges have been performed. Further, to the best of our knowledge, no ex vivo study has systematically investigated phonatory parameters from high-speed, audio and subglottal pressure data with irregular oscillations. To remedy this, the present study analyzes experiments with sustained phonation in 11 ex vivo rabbit larynges for 51 conditions of disordered vocal fold dynamics. (1) The results of this study support previous findings on non-disordered data, that the stronger the glottal closure insufficiency is during phonation, the worse the phonatory characteristics are; (2) aperiodic oscillations showed worse phonatory results than subharmonic oscillations; (3) in the presence of both types of irregular vibrations, the voice quality (i.e., cepstral peak prominence) of the audio and subglottal signal greatly deteriorated compared to normal/periodic vibrations. In summary, our results suggest that the presence of both types of irregular vibration have a major impact on voice quality and should be considered along with glottal closure measures in medical diagnosis and treatment.

Emulating the perceptual capabilities of a human evaluator to map the GRB scale for the assessment of voice disorders

This paper presents the design of an automatic voice quality analysis system for the assessment of voice pathologies, which emulates the perceptual capabilities of a human evaluator according the GRB scale. For this purpose, a novel methodology based on multiple sets of characteristics, ordinal classification and Gaussian regression is proposed. In particular, a reduced subset of characteristics is identified, and the regressor is used to convert the discrete perceptual scale to a continuum, more in agreement to the nature of the problem under study. The robustness of the system is evaluated in several cross-dataset experiments. Similarly, a clinical evaluation of the predictions provided by the system is carried out. Results indicate that the proposed methodology is proficient in modelling the perceptual capabilities of the human evaluator. They also show that it is possible to extend the GRB scale to a continuum through regression techniques while maintaining the consistency of the results. On average, the deviation between the labels assessed by the expert and the ones provided by the system is of about 0.5 units (in a scale from 0 to 3) for G and B, and of 0.7 units for R. Similarly, the deviation of the labels predicted by the system in the clinical assessment trials is about 0.3 units for G, 0.4 units for B, and 0.5 units for R.

Exploring the feasibility of the combination of acoustic voice quality index and glottal function index for voice pathology screening.

The aim of this study was to explore the diagnostic value of the combination of Acoustic Voice Quality Index (AVQI) and Glottal Function Index (GFI) as a screening tool for voice disorders, and to compare the AVQI measurements obtained using oral and smartphone (SP) microphones. A study group consisted of 183 adult individuals including 86 subjects with normal voice and 97 patients with pathological voice. Voice recordings were performed simultaneously with an oral AKG Perception 220 and SP iPhone 6s microphones. To evaluate the diagnostic accuracy differentiating normal and pathological voice, the receiver-operating characteristic statistics [area under curve (AUC), positive and negative likelihood ratios (LR+ and LR-)], and correct classification rate (CCR) were used. The AVQI cut-off scores of 3.31 for oral and 3.32 for SP microphones were associated with very good test accuracy (AUC = 0.857 and AUC = 0.818), resulting in balance between sensitivity and specificity (70.0% vs 86.0% and 70% vs 87.0%). The CCR reached 78%. The combined AVQI and GFI cut-off scores of 6.65 for oral and 7.1 for SP microphones corresponded with excellent test accuracy (AUC = 0.976 and AUC = 0.965) and sensitivity and specificity (93.0% vs 93.0% and 91.0% vs 94%). Very respectable levels of LR+ and LR- both for oral microphone (13.3 and 0.08) and for SP microphone (15.6 and 0.10) voice recordings were achieved. CCRs of 93% and 92% confirmed the results of ROC statistics. Combination of AVQI and GFI measurements significantly improved diagnostic accuracy in differentiating normal vs pathological voice.

Perceptual Judgment of Voice Quality in Nondysphonic French Speakers: Effect of Task-, Speaker- and Listener-Related Variables

Several perceptual scales have been developed to assess voice quality in dysphonic voices, among which the Grade Roughness Breathiness Asthenia Strain and a Rate of Dysphonia scale is probably the most frequently used. However, this clinical tool has not been properly validated with a normophonic population yet. The aim of the present study was to provide a first set of reference data gathered from a normal population, to serve as a basis of comparison for vocologists and laryngologists working with French-speaking patients. A second goal was to investigate the influence on this normal voice dataset, of variables known to affect perceptual judgments of pathological voice. Sustained vowels and sentences produced by 80 healthy, normophonic French native speakers were perceptually assessed by a panel of 18 raters (nine students, nine experts) using the Grade Roughness Breathiness Asthenia Strain and a Rate of Dysphonia scale. The average overall grade was close to 1 on the (0 to 3) scale, questioning the notion of "normal" voice as opposed to dysphonic voice. Rating reliability as well as perceptual scores were affected by task-, speaker-, and listener-related factors: speech stimuli led to better rating reliability and were judged less severely than voice stimuli; experts were slightly more reliable and less severe than students; older speakers were unanimously considered as more dysphonic. Multiple interactions between these factors were observed, confirming the multidimensional nature of voice quality.

Implementation of voice pathology detection system using feature selection

Implementation of voice pathology detection system using feature selection

On the design of automatic voice condition analysis systems. Part I: Review of concepts and an insight to the state of the art

This is the first of a two-part series devoted to review the current state of the art of automatic voice condition analysis systems. The goal of this paper is to provide to the scientific community and to newly comers to the field of automatic voice condition analysis a resource that presents introductory concepts, a categorisation of different aspects of voice pathology and a systematic literature review that describes the methodologies and methods that are mostly used in these systems. To this end, pathological voice is firstly described in terms of perceptual characteristics and its relationship with physiological phenomena. Then, a prototypical automatic voice condition analysis system is described, discussing each one of its constituting parts and presenting an in-depth literature review about the methodologies that are typically employed. Finally, a discussion about some variability factors that affect the performance of these systems is presented.

Diagnostic accuracy of the Smoothed Cepstral Peak Prominence (CPPS) in the detection of dysphonia in the Spanish language

The smoothed cepstral peak prominence (CPPS) is an acoustic measure that can be calculated in both sustained vowels and continuous speech. The goal of this work is to find out the diagnostic accuracy of CPPS in the detection of dysphonia in Spanish.
 In this study 136 subjects with dysphonia and 47 healthy subjects participated. For each subject a sustained vowel and the reading of three phonetically balanced sentences were recorded. The CPPS was calculated with Praat using its default configuration (configuration 1), and also with the one used in the calculation of the Acoustic Voice Quality Index (configuration 2). Five experts perceptively assessed the voice of the subjects in the sample by means of the GRABS scale.
 The CPPS has a great power of discrimination between the normal and the pathological voice, whether it is calculated from the sustained vowel /a/ (AROC[config. 1] = 0.863 and AROC[config. 2] = 0.841) or whether it is calculated from the sentences (AROC[config. 1] = 0.884 and AROC[config. 2] = 0.866).
 The results confirm that CPPS is a valid acoustic measurement to detect dysphonia in the Spanish language.

Detection of extra pulses in synthesized glottal area waveforms of dysphonic voices

Background and objectivesThe description of production kinematics of dysphonic voices plays an important role in the clinical care of voice disorders. However, high-speed videolaryngoscopy is not routinely used in clinical practice, partly because there is a lack of diagnostic markers that may be obtained from high-speed videos automatically. Aim of the study is to propose and test a procedure that automatically detects extra pulses, which may occur in voiced source signals of pathological voices in addition to cyclic pulses. Material and methodsGlottal area waveforms (GAW) are synthesized and used to test a detector for extra pulses. Regarding synthesis, for each GAW a cyclic pulse train is mixed with an extra pulse train, and additive noise. The cyclic pulse trains are varied across GAWs in terms of fundamental frequency, pulse shape, and modulation noise, i.e., jitter and shimmer. The extra pulse trains are varied across GAWs in terms of the height of the extra pulses, and their rates of occurrence. The energy level of the additive noise is also varied. Regarding detection, first, the fundamental frequency is estimated jointly with the cyclic pulse train waveform, second, the modulation noise is estimated, and finally the extra pulse train waveform is estimated. Two versions of the detector are compared, i.e., one that parameterizes the shapes of the cyclic pulses, and one that uses unparameterized pulse shape estimates. Two corpora are used for testing, i.e., one with 100 GAWs containing random extra pulses, and one with 25 GAWs containing extra pulses in the closed phases of each glottal phase representing subharmonic voices. Results and discussionWith pulse shape parameterization (PSP) a maximum mean accuracy of 88.3% is achieved when detecting random extra pulses. Without PSP, the maximum mean accuracy reduces to 82.9%. Detection performance decreases if the energy level of additive noise is higher than −25 dB with respect to the energy of the cyclic pulse train, and if the irregularity strength exceeds 0.1. For bicyclic, i.e., subharmonic voices, the approach fails without PSP, whereas with PSP, a mean sensitivity of 87.4% is achieved for subharmonic voices. ConclusionA synthesizer for GAWs containing extra pulses, and a detector for extra pulses are proposed. With PSP, favorable detector performance is observed for not too high levels of additive noise and irregularity strengths. In signals with high noise levels, the detector without PSP outperforms the other one. Detection of extra pulses fails if irregularity strength is large. For subharmonic voices PSP must be used.

A Pathological Multi-Vowels Recognition Algorithm Based on LSP Feature

At present, pathological voice recognition is mainly based on the classification of pathological voice. However, almost all the researches are based on the single vowel \a samples, but few on multi-vowels. In addition, the current researches on multi-vowels recognition are mainly for normal voices, which are unsuitable for the speech recognition of normal and pathological multi-vowels simultaneously. This paper concentrates on developing an accurate and robust feature called enhanced-bark line spectrum pair (E-BLSP) to detect and classify normal and pathological multi-vowels. We explore the impact of E-BLSP feature on recognition performance and propose an effective method based on the combination of three features including E-BLSP for pathological and normal multi-vowels. In this paper, first LSP and difference of adjacent LSP (DAL) features of a vowel are extracted. Then LSP feature is warped at bark domain to get bark line spectrum pair (BLSP). In addition, then E-BLSP feature is calculated by adjusting BLSP using DAL feature. Finally, the adjusted E-BLSP feature and other two traditional features, including linear prediction cepstrum coefficient (LPCC) and mel-frequency cepstrum coefficients (MFCC) are applied to support vector machine (SVM) and deep neural network (DNN) classifiers to explore the classification performance of single feature and feature combinations for pathological and normal vowels /a/, /i/ and /u/. The results show that the highest achieved accuracies for DNN and SVM network are 98.6190% and 96.2693%, while the largest achieved area under curves (AUC) are 0.9925 and 0.9868, correspondingly with the combination of three features including LPCC, MFCC, and E-BLSP.

Leveraging Artificial Intelligence to Improve Voice Disorder Identification Through the Use of a Reliable Mobile App

The evolution of the Internet of Things, cloud computing and wireless communication has contributed to an advance in the interconnectivity, efficiency and data accessibility in smart cities, improving environmental sustainability, quality of life and well-being, knowledge and intellectual capital. In this scenario, the satisfaction of security and privacy requirements to preserve data integrity, confidentiality and authentication is of fundamental importance. In particular, this is essential in the healthcare sector, where health-related data are considered sensitive information able to reveal confidential details about the subject. In this regard, to limit the possibility of security attacks or privacy violations, we present a reliable mobile voice disorder detection system capable of distinguishing between healthy and pathological voices by using a machine learning algorithm. This latter is totally embedded in the mobile application, so it is able to classify the voice without the necessity of transmitting user data to or storing user data on any server. A Boosted Trees algorithm was used as the classifier, opportunely trained and validated on a dataset composed of 2003 voices. The most frequently considered acoustic parameters constituted the inputs of the classifier, estimated and analyzed in real time by the mobile application.

Outliers Treatment to Improve the Recognition of Voice Pathologies

Abstract In some of the processes used in data analysis, such as the recognition of pathologies and pathological subjects, the presence of anomalous instances in the dataset is an unfavorable situation that can lead to misleading results. This article presents a function that implements the identification of anomalies in dataset using the boxplot and standard deviation methods. Also was used the filling technique to treat these anomalies, in which the anomalous point value were substituted by a limit value determined by the boxplot or standard deviation methods. To improve the outliers methods some normalization processes based on the z-score, logarithmic and squared root methodologies were experimented. These outliers treatment were applied to the dataset used in the recognition of vocal pathologies (dysphonia, chronic laryngitis and vocal cords paralysis vs control), performed by a MLP and LSTM neural networks. After the experiments, both the standard deviation and the boxplot methods with z-score normalization showed very useful for pre-processing the dataset for voice pathologies recognition. The accuracy was improved between 3 and 13 points in percentage.

A Solution to the Security Authentication Problem in Smart Houses Based on Speech

Voice-operated smart houses often face security problems when there is pathology in the voice of an individual. If an error occurs during the process of authentication, then the system should detect whether this error is due to pathology or someone is trying to deceive the system. In the case of successful pathology detection, the system offers a second option of the pin code. Pathology detection is one of the major components in this research work. Voice pathology is the disorder of the vocal fold, which is normally diagnosed in people working in different areas, such as education, courtroom, etc. Many researchers have proposed different voice pathology detection systems (VPDS) where they used a common feature extraction technique, i.e., Mel Frequency cepstral coefficient (MFCC), or compared it with their proposed technique and produced results that are different in almost every case. In this paper, we propose a system that is based on MFCC features fused with pitch and it achieves 99.97% accuracy of pathology detection for the MEEI data set. This voice pathology detection accuracy is better as compared to most of the VPDS available for MEEI.

Relation of RBH Auditory-Perceptual Scale to Acoustic and Electroglottographic Voice Analysis in Children With Vocal Nodules

The aim of this paper was to present an analysis of the feasibility of voice quality prediction on the roughness, breathiness, hoarseness (RBH) scale for children with vocal nodules on the basis of both acoustic parameters and electroglottographic (EGG) examination. The first step to achieve this goal involved the creation of a dedicated database, Voice Pathology Analysis Database (VPADB), containing voice recordings from patients, the EGG signal, medical diagnosis, and the classification of voice quality on the above-mentioned scale. The database also contains data concerning the patients' age and sex. The next step involved performing statistical analyses to test the relationship between the values of objective parameters, such as peak slope and normalized amplitude quotient, and the classification of voice quality. The study made use of voice recordings of 57 patients with vocal nodules and 37 healthy individuals. The RBH classification was carried out by two independent voice specialists. It was found that speech signal parameters can be used to predict expert evaluation with regard to roughness and hoarseness.

Predicting Normal and Pathological Voice using a Cepstral Based Acoustic Index in Sustained Vowels versus Connected Speech

배경 및 목적: 본 연구에서는 모음연장발성과 연결발화에서 음성장애 진단을 위한 기준값을 정희하고, 두 음성표본 중 음성장애를 분석하기 위한 민감한 표본이 무엇인지 알아보았다. 방법: 74명의 정상화자와 214명의 음성장애 환자를 대상으로 모음연장발성과 연결발 화의 캡스트럼 및 스펙트럼 변수를 측정하였다. 음성장애 유무와 음성표본에 따라 각 측정치를...

Principais patologias laríngeas em professores

Introdução: Os docentes estão expostos a diversos fatores de riscos ocupacionais, e acredita-se que qualidade vocal não está relacionada apenas com ausência de alteração, mas sim com bem-estar geral do indivíduo. Existem vários fatores que contribuem para o aparecimento de problemas vocais nos professores como: o uso inadequado da voz, competição vocal, sobrecarga vocal, hábitos inadequados como exemplo (cigarro e álcool). Objetivo: O presente estudo teve como objetivo analisar as principais alterações vocais em professores e consistiu em uma pesquisa bibliográfica, descritiva de caráter qualitativo. Método: Realizou-se um levantamento da literatura nacional e internacional, publicada nos idiomas Inglês, Espanhol, Português entre 2007 e 2017. Foram selecionados por meio das bases de dados “Biblioteca Virtual em Saúde” (BVS) e “Scientific Electronic Library” (SciELO), utilizando-se os descritores, professores, patologias da voz, saúde vocal e prevenção. Resultados: Observou-se que as principais patologias encontradas foram os nódulos, pólipos, edema de Reinke, cisto e sulco vocal. A rouquidão é o principal sintoma descrito pelos docentes, mas é preciso se preocupar com os demais sintomas sendo eles: fadiga vocal, falhas na voz, falta de ar, ardência na garganta, esforço para falar. Conclusão: Verificou-se na literatura estudada a sugestão de adotar uma política pública de saúde com o objetivo de educar e conscientizar os docentes quanto a noções de higiene vocal, estratégias de comunicação, economia vocal e valorização da profissão, que são ferramentas necessárias para diminuir os índices elevados de docentes com algum tipo de alteração vocal.

Cervical Spine Disability in Correlation with Subjective Voice Handicap in Patients With Voice Disorders: A Retrospective Analysis

Neck muscle dysfunction has been considered as risk factor or consequence of voice disorders. This study investigates the correlation between neck and voice-related subjective symptoms in patients with voice disorders. Retrospective case-control study. 100 adult patients (59 female and 41 male) over 18 years with a mean age of 50.01 years (SD 16, range 24-87), of which 68 were under 60 years, were included. 50 patients had organic voice pathologies and 50 functional dysphonia. Outcome measures were the Voice Handicap Index 9 international (VHI-9i) and the Neck Disability Index (NDI-G). Spearman rank order coefficient was applied to determine the correlation between overall and single item VHI and NDI results. Subanalyses were done for functional vs. organic disorder, gender and age ± 60 years. Mean overall VHI-9i (13.93, SD = 7.81, range = 0-31) and mean NDI-G (6.07, SD = 7.71, range = 0-43) showed a significant mild correlation (rs = 0.220, P = 0.02). Split into subgroups the relation was stronger in patients with organic pathologies (rs = 0.297, P = 0.03), but not significant in functional disorders (r = 0.148, P = 0.30). There was a moderate relation in men (rs = 0.317, P = 0.04). Single item correlation was highest between VHI-9i item P4 (physiological) and NDI-G item reading (cognitive functioning) (rs = 0.480, P = 0.002). Specifically patients with organic voice disorders showed increased voice symptoms with the presence of neck dysfunction. This indicates a risk for a functional imbalance of the muscles surrounding the larynx, which in extreme cases may hinder functional voice rehabilitation even after phonosurgery. Therefore, neck dysfunction should be considered in voice diagnostics.

Comparison of Two Versions of the Acoustic Voice Quality Index for Quantification of Dysphonia Severity

The acoustic voice quality index (AVQI) is a specific acoustic indicator designed to objectively estimate dysphonia severity and measure the values of acoustic parameters based on the diagnostic category. This study compared the performance of two AVQI versions (2.02 and 3.01, v2 and v3) and PraatCPPS using a voice sample of Korean population. Voice samples for sustained vowel and connected speech were elicited from 2257 patients across 14 diagnostic categories. Auditory-perceptual (A-P) assessments of dysphonia severity were compared to acoustic parameters of severity derived from two versions of the AVQI (v2 and v3) as well as the PraatCPPS. The AVQI-estimated severity (v2 and v3) and PraatCPPS severity for concatenated voice samples strongly correlated with each other and were significantly associated with A-P ratings. The AVQI (v2 and v3) and PraatCPPS showed high reliability in differentiating between pathological voice disorders. The AVQI (v2 and v3) and PraatCPPS were strongly correlated with the A-P ratings and provided valid estimates of dysphonia severity. However, the associations of the A-P ratings with the AVQIv2 were significantly stronger than those with the AVQIv3 and PraatCPPS, suggesting that the V2 outperformed the V3 and PraatCPPS.