Discriminant Capacity of Acoustic, Perceptual, and Vocal Self: The Effects of Vocal Demands

Disease Delineation for Multiple Sclerosis, Friedreich Ataxia, and Healthy Controls Using Supervised Machine Learning on Speech Acoustics

Neurodegenerative disease often affects speech. Speech acoustics can be used as objective clinical markers of pathology. Previous investigations of pathological speech have primarily compared controls with one specific condition and excluded comorbidities. We broaden the utility of speech markers by examining how multiple acoustic features can delineate diseases. We used supervised machine learning with gradient boosting (CatBoost) to delineate healthy speech from speech of people with multiple sclerosis or Friedreich ataxia. Participants performed a diadochokinetic task where they repeated alternating syllables. We subjected 74 spectral and temporal prosodic features from the speech recordings to machine learning. Results showed that Friedreich ataxia, multiple sclerosis and healthy controls were all identified with high accuracy (over 82%). Twenty-one acoustic features were strong markers of neurodegenerative diseases, falling under the categories of spectral qualia, spectral power, and speech rate. We demonstrated that speech markers can delineate neurodegenerative diseases and distinguish healthy speech from pathological speech with high accuracy. Findings emphasize the importance of examining speech outcomes when assessing indicators of neurodegenerative disease. We propose large-scale initiatives to broaden the scope for differentiating other neurological diseases and affective disorders.

Characterization of Primary Muscle Tension Dysphonia Using Acoustic and Aerodynamic Voice Metrics

OBJECTIVES/HYPOTHESIS

The objectives of this study were to (1) identify optimal clusters of 15 standard acoustic and aerodynamic voice metrics recommended by the American Speech-Language-Hearing Association (ASHA) to improve characterization of patients with primary muscle tension dysphonia (pMTD) and (2) identify combinations of these 15 metrics that could differentiate pMTD from other types of voice disorders.

STUDY DESIGN

Retrospective multiparametric METHODS: Random forest modeling, independent t-tests, logistic regression, and affinity propagation clustering were implemented on a retrospective dataset of 15 acoustic and aerodynamic metrics.

RESULTS

Ten percent of patients seen at the New York University (NYU) Voice Center over two years met the study criteria for pMTD (92 out of 983 patients), with 65 patients with pMTD and 701 of non-pMTD patients with complete data across all 15 acoustic and aerodynamic voice metrics. PCA plots and affinity propagation clustering demonstrated substantial overlap between the two groups on these parameters. The highest ranked parameters by level of importance with random forest models-(1) mean airflow during voicing (L/sec), (2) mean SPL during voicing (dB), (3) mean peak air pressure (cmH2O), (4) highest F0 (Hz), and (5) CPP mean vowel (dB)-accounted for only 65% of variance. T-tests showed three of these parameters-(1) CPP mean vowel (dB), (2) highest F0 (Hz), and (3) mean peak air pressure (cmH2O)-were statistically significant; however, the log2-fold change for each parameter was minimal.

CONCLUSION

Computational models and multivariate statistical testing on 15 acoustic and aerodynamic voice metrics were unable to adequately characterize pMTD and determine differences between the two groups (pMTD and non-pMTD). Further validation of these metrics is needed with voice elicitation tasks that target physiological challenges to the vocal system from baseline vocal acoustic and aerodynamic ouput. Future work should also place greater focus on validating metrics of physiological correlates (eg, neuromuscular processes, laryngeal-respiratory kinematics) across the vocal subsystems over traditional vocal output measures (eg, acoustics, aerodynamics) for patients with pMTD.

LEVEL OF EVIDENCE

II.

Systematic Review of Literature on Vocal Demand Response: Understanding Physiology, Measurements, and Associated Factors

PURPOSE

Considering the conceptual migration from vocal load and vocal loading to vocal demand and vocal demand response, this review of literature aimed to identify physiological explanations, reported measurements, and associated factors (vocal demands) reported in the literature when considering the phonatory response to a vocal demand.

METHODS

A systematic review of literature, following the PRISMA Statement, was conducted using Web of Science, PubMed, Scopus, and ScienceDirect. Data were analyzed and presented in two parts. First, a bibliometric analysis, co-occurrence analysis, and content analysis were performed. Three criteria that got article inclusion were defined: (1) written in English, Spanish, and Portuguese; (2) published between 2009 and 2021; and (3) focused on vocal load and loading, vocal demand response, and voice assessment parameters. A total of 54 publications met the criteria and were included in this review. The second part included a conceptual framework based on the content analysis of three aspects of vocal demand response: (1) physiological explanations, (2) reported measurements, and (3) vocal demands.

RESULTS AND CONCLUSION

As would be expected since vocal demand response is a relatively new term and not yet commonly used in literature when discussing way that the speakers respond to communicative scenarios, most of the studies reviewed (both historical and recent) still use the term of vocal load and vocal loading. Although there is a broad variety of literature discussing a wide range of vocal demands and voice parameters used to characterize the vocal demand response, results show that there is consistency across the studies. While vocal demand response is unique and intrinsic to the talker, associated factors that contribute to this response include both internal talker and external talker factors. Internal factors include muscle stiffness, viscosity in the phonatory system, vocal fold tissue damage, elevated sound pressure levels during occupational voice demands, extended periods of voice use, suboptimal body posture, difficulties in breathing technique, and sleep disturbances. Associated external factors include the working environment (noise, acoustics, temperature, humidity). In conclusion, although vocal demand response is intrinsic to the speaker, the speaker's response is affected by external vocal demands. However, due to the wide methods to evaluate vocal demand response, it has been difficult to establish its contribution to voice disorders in the general population and, specifically, among occupational voice users. This literature review identified commonly reported parameters and factors that may help clinicians and researchers define vocal demand response.

Self-reported Vocal Demands, Perceptions, and Knowledge of Occupational Voice Users in South Africa

OBJECTIVES

The study aimed to describe daily vocal demands, perceptions and knowledge as reported by occupational voice users.

STUDY DESIGN

A descriptive, cross-sectional research design was employed.

METHODS

A survey on vocal demands, perceptions and knowledge was distributed to 102 occupational voice users via a snowball sampling technique.

RESULTS

Slightly more than half of the participants (55%) reported using their voice for work 36.5 hours a week on average (SD = 15.5, range: 33-40). Participants reported that, on average, their daily voice use is 6.3 hours (SD = 2.7) for work and the majority (81%) reported a decrease in voice quality after work; Three-quarters (75%) also reported vocal fatigue at the end of day. Approximately one-third (33%) reported being exposed to environments where they are expected to shout, scream or cheer loudly. More than half of the participants (61%) reported that they have previously received vocal health education but 40% reported that they felt this training was insufficient. High vocal demands are significantly correlated to an increase in perceived vocal handicap rs = 0.242; (P = 0.018), tiredness of voice rs = 0.270; (P = 0.008), physical discomfort rs = 0.217; (P = 0.038) as well as how occupational voice users experience improvement of symptoms with rest rs = -0.356 (P < 0.001). Other risk factors highlighted by occupational voice users are the ingestion of liquid caffeine, alcohol, and carbonated drinks, smoking or the presence of chronic cough, chronic laryngitis, and gastroesophageal reflux disease.

CONCLUSIONS

Occupational voice users are subjected to high daily vocal demands that can be seen to be linked to vocal fatigue, changes in voice quality and vocal symptoms experienced. It is essential that occupational voice users as well as treating clinicians are aware of certain significant predictors of both vocal handicap and vocal fatigue. These findings provide insights for developing strategies for training and cultivating vocal health consciousness and preventive voice care initiatives that focus on occupational voice users in South Africa.

Vocal dose in teachers: correlation with dysphonia

Teachers are professionals with high prevalence of dysphonia, whose main risk factors are the large work hours in classrooms with the presence of background noise. The purpose of the study was to calculate the phonation time and the cycle dose of teachers with dysphonia and teachers without voice disorders during the class. There were two groups analyzed: five teachers with functional dysphonia were the first group and five teachers without voice disorders were the second group. For the data was used the VoxLog® dosimeter and the parameters were: intensity; fundamental frequency; phonation time and cycle dose. The statistical analysis used ANOVA, Student's T-test, and Kruskal-Wallis test. Dysphonic teachers showed major values of phonation time and cycle dose compared with teachers without voice disorders. The dysphonia is related to extended period of speech time and greater exposure of the tissue of the vocal fold to phonotrauma.

Dysphonia risk screening protocol

OBJECTIVE

To propose and test the applicability of a dysphonia risk screening protocol with score calculation in individuals with and without dysphonia.

METHOD

This descriptive cross-sectional study included 365 individuals (41 children, 142 adult women, 91 adult men and 91 seniors) divided into a dysphonic group and a non-dysphonic group. The protocol consisted of 18 questions and a score was calculated using a 10-cm visual analog scale. The measured value on the visual analog scale was added to the overall score, along with other partial scores. Speech samples allowed for analysis/assessment of the overall degree of vocal deviation and initial definition of the respective groups and after six months, the separation of the groups was confirmed using an acoustic analysis.

RESULTS

The mean total scores were different between the groups in all samples. Values ranged between 37.0 and 57.85 in the dysphonic group and between 12.95 and 19.28 in the non-dysphonic group, with overall means of 46.09 and 15.55, respectively. High sensitivity and specificity were demonstrated when discriminating between the groups with the following cut-off points: 22.50 (children), 29.25 (adult women), 22.75 (adult men), and 27.10 (seniors).

CONCLUSION

The protocol demonstrated high sensitivity and specificity in differentiating groups of individuals with and without dysphonia in different sample groups and is thus an effective instrument for use in voice clinics.