Artificial Intelligence Procedure for the Screening of Genetic Syndromes Based on Voice Characteristics.

Perceptual and statistical evidence has highlighted voice characteristics of individuals affected by genetic syndromes that differ from those of normophonic subjects. In this paper, we propose a procedure for systematically collecting such pathological voices and developing AI-based automated tools to support differential diagnosis. Guidelines on the most appropriate recording devices, vocal tasks, and acoustical parameters are provided to simplify, speed up, and make the whole procedure homogeneous and reproducible. The proposed procedure was applied to a group of 56 subjects affected by Costello syndrome (CS), Down syndrome (DS), Noonan syndrome (NS), and Smith-Magenis syndrome (SMS). The entire database was divided into three groups: pediatric subjects (PS; individuals < 12 years of age), female adults (FA), and male adults (MA). In line with the literature results, the Kruskal-Wallis test and post hoc analysis with Dunn-Bonferroni test revealed several significant differences in the acoustical features not only between healthy subjects and patients but also between syndromes within the PS, FA, and MA groups. Machine learning provided a k-nearest-neighbor classifier with 86% accuracy for the PS group, a support vector machine (SVM) model with 77% accuracy for the FA group, and an SVM model with 84% accuracy for the MA group. These preliminary results suggest that the proposed method based on acoustical analysis and AI could be useful for an effective, non-invasive automatic characterization of genetic syndromes. In addition, clinicians could benefit in the case of genetic syndromes that are extremely rare or present multiple variants and facial phenotypes.

Machine Learning Assessment of Spasmodic Dysphonia Based on Acoustical and Perceptual Parameters.

Adductor spasmodic dysphonia is a type of adult-onset focal dystonia characterized by involuntary spasms of laryngeal muscles. This paper applied machine learning techniques for the severity assessment of spasmodic dysphonia. To this aim, 7 perceptual indices and 48 acoustical parameters were estimated from the Italian word /a'jwɔle/ emitted by 28 female patients, manually segmented from a standardized sentence and used as features in two classification experiments. Subjects were divided into three severity classes (mild, moderate, severe) on the basis of the G (grade) score of the GRB scale. The first aim was that of finding relationships between perceptual and objective measures with the Local Interpretable Model-Agnostic Explanations method. Then, the development of a diagnostic tool for adductor spasmodic dysphonia severity assessment was investigated. Reliable relationships between G; R (Roughness); B (Breathiness); Spasmodicity; and the acoustical parameters: voiced percentage, F2 median, and F1 median were found. After data scaling, Bayesian hyperparameter optimization, and leave-one-out cross-validation, a k-nearest neighbors model provided 89% accuracy in distinguishing patients among the three severity classes. The proposed methods highlighted the best acoustical parameters that could be used jointly with GRB indices to support the perceptual evaluation of spasmodic dysphonia and provide a tool to help severity assessment of spasmodic dysphonia.

Aggregate Channel Features for newborn face detection in Neonatal Intensive Care Units.

An efficient face detector could be very helpful to point out possible neurological dysfunctions such as seizure events in Neonatal Intensive Care Units. However, its development is still challenging because large public datasets of newborns' faces are missing. Over the years several studies introduced semi-automatic approaches. This study proposes a fully automated face detector for newborns in Neonatal Intensive Care Units, based on the Aggregate Channel Feature algorithm. The developed method is tested on a dataset of video recordings from 42 full-term newborns collected at the Neuro-physiopathology and Neonatology Clinical Units, AOU Careggi, Firenze, Italy. The proposed system showed promising results, giving (mean ± standard error): log-Average Miss Rate = 0.47 ± 0.05 and Average Precision Recall = 0.61 ± 0.05. Moreover, achieved results highlighted interesting differences between newborns without seizures, newborns with electro-clinical seizures, and newborns with electrographic-only seizures. For both metrics statistically significant differences were found between patients with electro-clinical seizures and the other two groups. Clinical Relevance- The proposed method, based on quantitative physio-pathological features of facial movements, is of clinical relevance as it could speed up pain or seizure assessment of newborns in Neonatal Intensive Care Units.

Analysis of Brain-Heart Interactions in newborns with and without seizures using the Convergent Cross Mapping approach.

In the last years, the characterization of brain-heart interactions (BHIs) in epilepsy has gained great interest. For some specific seizures there is still a lack of information about the mechanisms occurring during or close to ictal events between the central nervous system (CNS) and the autonomic nervous system (ANS). This is the case for neonatal seizures, one of the most common neurological emergencies in the first days of life. This paper evaluates possible differences in BHIs between newborns with seizures and seizure-free ones. We applied convergent cross mapping approaches to a cohort of 52 newborns from a public dataset. Preliminary results show that newborns with seizures have a lower degree of interaction between the CNS and the ANS than seizure-free ones (Mann-Whitney test: p-value <0.05). These results are of clinical relevance for future BHI-based approaches to better understand the neural mechanisms behind neonatal seizures. Clinical Relevance- The study of BHIs in newborns with seizures might be helpful to better characterize the disorder or the aetiologies behind ictal events. Moreover, BHI approaches may confirm the involvement of the ANS during or close to a neonatal seizure event.

Heart Rate Variability Analysis for Seizure Detection in Neonatal Intensive Care Units.

In Neonatal Intensive Care Units (NICUs), the early detection of neonatal seizures is of utmost importance for a timely clinical intervention. Over the years, several neonatal seizure detection systems were proposed to detect neonatal seizures automatically and speed up seizure diagnosis, most based on the EEG signal analysis. Recently, research has focused on other possible seizure markers, such as electrocardiography (ECG). This work proposes an ECG-based NSD system to investigate the usefulness of heart rate variability (HRV) analysis to detect neonatal seizures in the NICUs. HRV analysis is performed considering time-domain, frequency-domain, entropy and multiscale entropy features. The performance is evaluated on a dataset of ECG signals from 51 full-term babies, 29 seizure-free. The proposed system gives results comparable to those reported in the literature: Area Under the Receiver Operating Characteristic Curve = 62%, Sensitivity = 47%, Specificity = 67%. Moreover, the system's performance is evaluated in a real clinical environment, inevitably affected by several artefacts. To the best of our knowledge, our study proposes for the first time a multi-feature ECG-based NSD system that also offers a comparative analysis between babies suffering from seizures and seizure-free ones.

Semioccluded Vocal Tract Exercises Improve Self-Perceived Voice Quality in Healthy Actors.

PURPOSE: Semi-occluded vocal tract exercises (SOVTE) have shown to lead to more effective and efficient vocal production for individuals with voice disorders and for singers. The aim of the present study is to investigate the effects of a 10-minute SOVTE warm-up protocol on the actors' voice. METHODS: Twenty-seven professional theater actors (16 females) without voice complaints were audio-recorded while reading aloud, with their acting voice, a short dramatic passage at four time points. Recordings were made: the day before the show, just before and soon after the warm-up protocol which was performed prior to the show and soon after the show. The voice quality was acoustically and auditory-perceptually evaluated and quantified at each time point by blinded raters. Self-assessment parameters anonymously collected pre and post exercising were also analyzed. RESULTS: No statistically significant differences on perceptual ratings and acoustic parameters were found between pre/post exercise sessions and males/females. A statistically significant improvement was detected in the self-assessment parameters concerning comfort of production, sonorousness, vocal clarity and power. CONCLUSIONS: Vocal warm-up with the described SOVTE protocol was effective in determining a self-perceived improvement in comfort of production, voice quality and power, although objective evidence was missing. This straightforward protocol could thus be beneficial if routinely utilized by professional actors to facilitate the voice performance.

A Generalized Linear Model for an ECG-based Neonatal Seizure Detector.

Seizures represent one of the most challenging issues of the neonatal period's neurological emergency. Due to the heterogeneity of etiologies and clinical characteristics, seizures recognition is tricky and time-consuming. Currently, the gold standard for seizure diagnosis is Electroencephalography (EEG), whose correct interpretation requires a highly specialized team. Thus, to speed up and facilitate the detection of ictal events, several EEG-based Neonatal Seizure Detectors (NSDs) have been proposed in the literature. Research is currently exploiting more simple and less invasive approaches, such as Electrocardiography (ECG). This work aims at developing an ECG-based NSD using a Generalized Linear Model with features extracted from Heart Rate Variability (HRV) measures as input. The method is validated on a public dataset of 52 subjects (33 with seizures and 19 seizure-free). Achieved encouraging results show 69% Concatenated Area Under the ROC Curve (AUCcc) for the automatic detection of windows with seizure events, confirming that HRV features can be useful to catch the cardio-regulatory system alterations due to neonatal seizure events, particularly those related to Hypoxic-Ischaemic Encephalopathies. Thus, results suggest the use of ECG-based NSDs in clinical practice, especially when a timely diagnosis is needed and EEG technologies are not readily available.Clinical Relevance- An ECG-based Neonatal Seizure Detector could be a valid support to speed up the diagnosis of neonatal seizures, especially when EEG technologies for infants' neurological assessment are not readily available.

Multiscale Entropy Analysis of Heart Rate Variability in Neonatal Patients with and without Seizures.

The complex physiological dynamics of neonatal seizures make their detection challenging. A timely diagnosis and treatment, especially in intensive care units, are essential for a better prognosis and the mitigation of possible adverse effects on the newborn's neurodevelopment. In the literature, several electroencephalographic (EEG) studies have been proposed for a parametric characterization of seizures or their detection by artificial intelligence techniques. At the same time, other sources than EEG, such as electrocardiography, have been investigated to evaluate the possible impact of neonatal seizures on the cardio-regulatory system. Heart rate variability (HRV) analysis is attracting great interest as a valuable tool in newborns applications, especially where EEG technologies are not easily available. This study investigated whether multiscale HRV entropy indexes could detect abnormal heart rate dynamics in newborns with seizures, especially during ictal events. Furthermore, entropy measures were analyzed to discriminate between newborns with seizures and seizure-free ones. A cohort of 52 patients (33 with seizures) from the Helsinki University Hospital public dataset has been evaluated. Multiscale sample and fuzzy entropy showed significant differences between the two groups (p-value < 0.05, Bonferroni multiple-comparison post hoc correction). Moreover, interictal activity showed significant differences between seizure and seizure-free patients (Mann-Whitney Test: p-value < 0.05). Therefore, our findings suggest that HRV multiscale entropy analysis could be a valuable pre-screening tool for the timely detection of seizure events in newborns.

Multiparametric EEG analysis of brain network dynamics during neonatal seizures.

BACKGROUND: One of the most challenging issues in paediatric neurology is the diagnosis of neonatal seizures, whose delayed treatment may affect the neurodevelopment of the newborn. Formulation of the correct diagnosis is conditioned by the high number of perceptually or automatically detected false positives. NEW METHOD: New methodologies are proposed to assess neonatal seizures trend over time. Our approach is based on the analysis of standardized trends of two properties of the brain network: the Synchronizabilty (S) and the degree of phase synchronicity given by the Circular Omega Complexity (COC). Qualitative and quantitative methods based on network dynamics allow differentiating seizure events from interictal periods and seizure-free patients. RESULTS: The methods were tested on a public dataset of labelled neonatal seizures. COC shows significant differences among seizure and non-seizure events (p-value <0.001, Cohen's d 0.86). Combining S and COC in standardized temporal instants provided a reliable description of the physiological behaviour of the brain's network during neonatal seizures. COMPARISON WITH EXISTING METHOD(S): Few of the existing network methods propose an operative way for carrying their analytical approach into the diagnostic process of neonatal seizures. Our methods offer a simple representation of brain network dynamics easily implementable and understandable also by less experienced staff. CONCLUSIONS: Our findings confirm the usefulness of the evaluation of brain network dynamics over time for a better understanding and interpretation of the complex mechanisms behind neonatal seizures. The proposed methods could also reliably support existing seizure detectors as a post-processing step in doubtful cases.

Automatic detection and sonification of nonmotor generalized onset epileptic seizures: Preliminary results.

Long-term video-EEG monitoring has improved diagnosis and treatment of epilepsy, especially in children. However, the amount of data neurophysiologists must analyze has grown remarkably. The main purpose of this paper is to provide a diagnostic support to speed up and ease EEG interpretation for a specific application concerning absence seizures, a type of non-motor generalized epileptic seizures. The proposed method consists of a pre-processing step where signals are filtered through the Stationary Wavelet Transform for the reduction of possible artefacts. Subsequently, a supervised automatic classification method is implemented for seizure detection, based on the Support Vector Machine Fine Gaussian method. Finally, a post-processing step is implemented in which spatial and temporal thresholds are defined for both online and offline application. In addition, a method that applies sonification techniques is developed. Sonification techniques could speed up the process of interpreting information, allowing rapid clinical intervention and a continuous monitoring of the event. The dataset consists of 30 EEG recordings performed in 24 children with absence seizures, clinically evaluated at the Meyer Children's Hospital in Firenze, Italy. The method shows encouraging results both in terms of balanced accuracy (about 96%) and latency times (1.25 s on average), which might make it suitable for online clinical trials. In fact, it was implemented in the perspective of a possible real-time application in clinical practice.

Multidimensional Assessment of the Effectiveness of Group Voice Therapy.

OBJECTIVE: Group voice therapy has been successfully used in patients with dysphonia, but there is little objectively documented evidence of its effects on voice quality and the self-perception of voice fatigue. The aim of this study was to investigate the effectiveness of group therapy in patients with functional dysphonia and minor anatomic vocal fold pathologies linked by appearance and history to voice use in an objective and multidimensional manner. STUDY DESIGN: This is a prospective longitudinal study. METHODS: Before and after treatment, 34 adult women underwent perceptual voice assessments using the grade of dysphonia, roughness, breathiness, asthenia, and strain (GRBAS) scale, Evaluation Vocale Assistée (EVA) system aerodynamic and acoustic assessments, and maximum phonation time measurements, and made subjective evaluations using the Voice Handicap Index. The pretreatment baseline values of the participants were obtained by means of two examinations separated by an interval of 1 week. The parameters belonging to each main dimension were clustered by means of Z-transformation, and the corresponding Z-scores were analyzed. RESULTS: Group therapy was associated with a statistically significant improvement in the perceptual (P = 0.008), acoustic (P = 0.040), aerodynamic (P = 0.009 and <0.001), and self-evaluation parameters (P = 0.011). CONCLUSIONS: Our findings provide evidence that group voice therapy can be associated with improvements in perceptual, acoustic, aerodynamic, and self-evaluated parameters in some patients with dysphonia. Controlled, randomized studies are needed in follow-up. This method of treatment may be a means of reducing the costs and waiting lists associated with rehabilitative treatment, and enhancing patients' motivation and compliance.

Maximal Ambient Noise Levels and Type of Voice Material Required for Valid Use of Smartphones in Clinical Voice Research.

PURPOSE: Smartphone technology provides new opportunities for recording standardized voice samples of patients and transmitting the audio files to the voice laboratory. This drastically improves the achievement of baseline designs, used in research on efficiency of voice treatments. However, the basic requirement is the suitability of smartphones for recording and digitizing pathologic voices (mainly characterized by period perturbations and noise) without significant distortion. In a previous article, this was tested using realistic synthesized deviant voice samples (/a:/) with three precisely known levels of jitter and of noise in all combinations. High correlations were found between jitter and noise to harmonics ratio measured in (1) recordings via smartphones, (2) direct microphone recordings, and (3) sound files generated by the synthesizer. In the present work, similar experiments were performed (1) in the presence of increasing levels of ambient noise and (2) using synthetic deviant voice samples (/a:/) as well as synthetic voice material simulating a deviant short voiced utterance (/aiuaiuaiu/). RESULTS: Ambient noise levels up to 50 dBA are acceptable. However, signal processing occurs in some smartphones, and this significantly affects estimates of jitter and noise to harmonics ratio when formant changes are introduced in analogy with running speech. The conclusion is that voice material must provisionally be limited to a sustained /a/.

Analysis of facial expressions in parkinson's disease through video-based automatic methods.

BACKGROUND: The automatic analysis of facial expressions is an evolving field that finds several clinical applications. One of these applications is the study of facial bradykinesia in Parkinson's disease (PD), which is a major motor sign of this neurodegenerative illness. Facial bradykinesia consists in the reduction/loss of facial movements and emotional facial expressions called hypomimia. NEW METHOD: In this work we propose an automatic method for studying facial expressions in PD patients relying on video-based METHODS: 17 Parkinsonian patients and 17 healthy control subjects were asked to show basic facial expressions, upon request of the clinician and after the imitation of a visual cue on a screen. Through an existing face tracker, the Euclidean distance of the facial model from a neutral baseline was computed in order to quantify the changes in facial expressivity during the tasks. Moreover, an automatic facial expressions recognition algorithm was trained in order to study how PD expressions differed from the standard expressions. RESULTS: Results show that control subjects reported on average higher distances than PD patients along the tasks. COMPARISON WITH EXISTING METHODS: This confirms that control subjects show larger movements during both posed and imitated facial expressions. Moreover, our results demonstrate that anger and disgust are the two most impaired expressions in PD patients. CONCLUSIONS: Contactless video-based systems can be important techniques for analyzing facial expressions also in rehabilitation, in particular speech therapy, where patients could get a definite advantage from a real-time feedback about the proper facial expressions/movements to perform.

Markerless Analysis of Articulatory Movements in Patients With Parkinson's Disease.

OBJECTIVES: A large percentage of patients with Parkinson's disease have hypokinetic dysarthria, exhibiting reduced peak velocities of jaw and lips during speech. This limitation implies a reduction of speech intelligibility for such patients. This work aims at testing a cost-effective markerless approach for assessing kinematic parameters of hypokinetic dysarthria. STUDY DESIGN: Kinematic parameters of the lips are calculated during a syllable repetition task from 14 Parkinsonian patients and 14 age-matched control subjects. METHODS: Combining color and depth frames provided by a depth sensor (Microsoft Kinect), we computed the three-dimensional coordinates of main facial points. The peak velocities and accelerations of the lower lip during a syllable repetition task are considered to compare the two groups. RESULTS: Results show that Parkinsonian patients exhibit reduced peak velocities of the lower lip, both during the opening and the closing phase of the mouth. In addition, peak values of acceleration are reduced in Parkinsonian patients, although with significant differences only in the opening phase with respect to healthy control subjects. CONCLUSIONS: The novel contribution of this work is the implementation of an entirely markerless technique capable to detect signs of hypokinetic dysarthria for the analysis of articulatory movements during speech. Although a large number of Parkinsonian patients have hypokinetic dysarthria, only a small percentage of them undergoes speech therapy to increase their articulatory movements. The system proposed here could be easily implemented in a home environment, thus, increasing the percentage of patients who can perform speech rehabilitation at home.

Application of Pattern Recognition Techniques to the Classification of Full-Term and Preterm Infant Cry.

OBJECTIVES: Scientific and clinical advances in perinatology and neonatology have enhanced the chances of survival of preterm and very low weight neonates. Infant cry analysis is a suitable noninvasive complementary tool to assess the neurologic state of infants particularly important in the case of preterm neonates. This article aims at exploiting differences between full-term and preterm infant cry with robust automatic acoustical analysis and data mining techniques. STUDY DESIGN: Twenty-two acoustical parameters are estimated in more than 3000 cry units from cry recordings of 28 full-term and 10 preterm newborns. METHODS: Feature extraction is performed through the BioVoice dedicated software tool, developed at the Biomedical Engineering Lab, University of Firenze, Italy. Classification and pattern recognition is based on genetic algorithms for the selection of the best attributes. Training is performed comparing four classifiers: Logistic Curve, Multilayer Perceptron, Support Vector Machine, and Random Forest and three different testing options: full training set, 10-fold cross-validation, and 66% split. RESULTS: Results show that the best feature set is made up by 10 parameters capable to assess differences between preterm and full-term newborns with about 87% of accuracy. Best results are obtained with the Random Forest method (receiver operating characteristic area, 0.94). CONCLUSIONS: These 10 cry features might convey important additional information to assist the clinical specialist in the diagnosis and follow-up of possible delays or disorders in the neurologic development due to premature birth in this extremely vulnerable population of patients. The proposed approach is a first step toward an automatic infant cry recognition system for fast and proper identification of risk in preterm babies.

Physiology and Acoustics of Inspiratory Phonation.

INTRODUCTION: Inspiratory phonation (IP) means phonating with inspiratory airflow. Some vocalists remarkably master this technique, to such an extent that it offers new dramatic, aesthetic, and functional possibilities in singing specific contemporary music. The present study aims to a better understanding of the physiological backgrounds of IP. MATERIAL AND METHODS: A total of 51 inhaling utterances were compared with 61 exhaling utterances in a professional soprano highly skilled in inhaling singing, by means of high-speed single-line scanning and advanced acoustic analysis. Ranges of intensity and Fo were kept similar. RESULTS: The main differences are: (1) an inversion of the mucosal wave, (2) a smaller closed quotient in IP, (3) a larger opening/closing quotient in IP with the additional difference that in IP, the quotient is larger than 1 (opening slower than closing), whereas it is less than 1 in expiratory mode (opening faster than closing), (4) a larger vocal-fold excursion in IP, (5) higher values of adaptive normalized noise energy in IP, and (6) a steeper slope of harmonic peaks in IP. However, jitter values are similar (within normal range), as well as damping ratios and central formant frequencies. The two voicing modes cannot be differentiated by blind listening. CONCLUSION: The basic physiological mechanisms are comparable in both voicing modes, although with specific differences. IP is actually to be considered as an "extended vocal technique," a term applied to vocalization in art music, which falls outside of traditional classical singing styles, but with remarkable possibilities in skilled vocalists.

Voice dosimetry and monitoring, with emphasis on professional voice diseases: Critical review and framework for future research.

Professional voice has become an important issue in the field of occupational health. Similarly, voice diseases related to occupations gain interest in insurance medicine, particularly within the frame of specific insurance systems for occupational diseases. Technological developments have made possible dosimetry of voice loading in the work-place, as well as long-term monitoring of relevant voice parameters during professional activities. A critical review is given, with focus on the specificity of occupational voice use and on the point of view of insurance medicine. Remaining questions and suggestions for future research are proposed.

Automatic Assessment of Acoustic Parameters of the Singing Voice: Application to Professional Western Operatic and Jazz Singers.

INTRODUCTION: The obvious perceptual differences between various singing styles like Western operatic and jazz rely on specific dissimilarities in vocal technique. The present study focuses on differences in vibrato acoustics and in singer's formant as analyzed by a novel software tool, named BioVoice, based on robust high-resolution and adaptive techniques that have proven its validity on synthetic voice signals. MATERIAL AND METHODS: A total of 48 professional singers were investigated (29 females; 19 males; 29 Western operatic; and 19 jazz). They were asked to sing "a cappella," but with artistic expression, a well-known musical phrase from Gershwin's Porgy and Bess, in their own style: either operatic or jazz. A specific sustained note was extracted for detailed vibrato analysis. Beside rate (s(-1)) and extent (cents), duration (seconds) and regularity were computed. Two new concepts are introduced: vibrato jitter and vibrato shimmer, by analogy with the traditional jitter and shimmer of voice signals. For the singer's formant, on the same sustained tone, the ratio of the acoustic energy in formants 1-2 to the energy in formants 3, 4, and 5 was automatically computed, providing a quality ratio (QR). RESULTS: Vibrato rates did not differ among groups. Extent was significantly larger in operatic singers, particularly females. Vibrato jitter and vibrato shimmer were significantly smaller in operatic singers. Duration of vibrato was also significantly longer in operatic singers. QR was significantly lower in male operatic singers. CONCLUSIONS: Some vibrato characteristics (extent, regularity, and duration) very clearly differentiate the Western operatic singing style from the jazz singing style. The singer's formant is typical of male operatic singers. The new software tool is well suited to provide useful feedback in a pedagogical context.

Automated tracking of quantitative parameters from single line scanning of vocal folds: a case study of the 'messa di voce' exercise.

This article presents a novel application of the 'single line scanning' of the vocal fold vibrations (kymography) in singing pedagogy, particularly in a specific technical voice exercise: the 'messa di voce'. It aims at giving the singer relevant and valid short-term feedback. A user-friendly automatic analysis program makes possible a precise, immediate quantification of the essential physiological parameters characterizing the changes in glottal impedance, concomitant with the progressive increase and decrease of the lung pressure. The data provided by the program show a strong correlation with the hand-made measurements. Additional measurements such as subglottic pressure and flow glottography by inverse filtering can be meaningfully correlated with the data obtained from the kymographic images.

Ambulatory phonation monitoring in a sample of 92 call center operators.

OBJECTIVES: The voice is a primary work tool for call center operators, but the main risk factors for voice disorders in this category have not yet been clarified. This study aimed to analyze the vocal behavior in call center operators and search for correlations between the daily voice dose and the self-perceived voice-related handicap. STUDY DESIGN: Prospective. SUBJECTS AND METHODS: Ninety-two call center operators (aged 24-50 years) underwent ambulatory phonation monitoring during a working day and were administered the Voice Handicap Index (VHI) questionnaire and a questionnaire concerning smoking habits, symptoms, and extrawork activities requiring intensive voice use. RESULTS: Mean percentage phonation time (PT) during work was 14.74% and ranged from 4% to 31%. There was a significant difference between the percentage PT in working time and in extrawork time; however, subjects with high percentage PT in working time maintained a high value also in extrawork time. The mean PT was 87.5 ± 35.8 minutes and was not correlated with age, gender, number of work hours, symptoms, extraprofessional voice use, and VHI scores. The mean amplitude was significantly higher in subjects with longer PT and higher pitch (P < 0.001). VHI score (median = 9) was slightly higher than in the general population but not related to the number of work hours, indicating that work time was not a critical factor in causing the perception of voice problems. CONCLUSION: Our study provides data about the voice behavior of a large cohort of call center operators and demonstrates that the number of work hours and the percentage PT are not statistically related to the perception of voice disturbances in this working category.