An Acoustic-Signal-Based Preventive Program for University Lecturers' Vocal Health

Machine Learning Models With Hyperparameter Optimization for Voice Pathology Classification on Saarbrücken Voice Database

Early diagnosis and referral are crucial in the treatment of voice disorders. Contemporary investigations have indicated the efficacy of voice pathology detection systems in significantly contributing to the evaluation of voice disorders, facilitating early diagnosis of such pathologies. These systems leverage machine learning methodologies, widely applied across diverse domains, and exhibit particular potential in the realm of voice pathology classification. However, machine learning models and performance metrics employed in these studies vary significantly, making it challenging to determine the optimal model for voice pathology classification. In this study, healthy and pathological voices were classified with state-of-the-art machine learning models, and the performance results of the models were compared. The voice samples employed in our research were sourced from the Saarbrücken Voice Database, a reputable German database. Feature extraction from voice signals was conducted using the Mel Frequency Cepstral Coefficients method. To assess and enhance the models' performance adequately, we employed hyperparameter optimization and implemented a 10-fold cross-validation approach. The outcomes revealed that the support vector machine model exhibited the highest accuracy, achieving 99.19% and 99.50% accuracies in the classification of male and female voice pathologies, respectively.

Altered Auditory Feedback in Teachers: A Preliminary Investigation

PURPOSE

Due to the elevated vocal risks of university professors and the possible relationship between auditory-motor integration and voice disorders, the current study was designed to explore the effects of altered auditory feedback via bone conduction on voice production measures in university professors.

METHODS

A total of 43 hours of voice recordings across 32 university classes were collected from two vocally healthy college professors through voice dosimetry. During their classes, the professors experienced either the real-time altered auditory feedback or a condition without altered auditory feedback. The voice dosimetry recordings from all classes were processed to calculate the sound pressure level values, fundamental frequency values, and the time dose. The effects of the altered auditory feedback conditions on these voice acoustic parameters were analyzed and compared with the conditions without altered auditory feedback.

RESULTS

The altered auditory feedback conditions resulted in significantly decreased sound pressure level values and time dose for both professors when comparing the altered auditory feedback conditions to the conditions without altered auditory feedback. The altered auditory feedback effects were larger for the male professor compared with the female professor. Additionally, the male professor demonstrated significantly decreased fundamental frequency values when comparing the altered auditory feedback conditions to the conditions without altered auditory feedback, while the female professor did not.

CONCLUSIONS

This study provides evidence that altered auditory feedback provided via bone conduction through an altered auditory feedback device resulted in statistically significant improvements in the voices of two college professors.

The Prevalence of Voice Disorders and the Related Factors in University Professors: A Systematic Review and Meta-Analysis

OBJECTIVE

This systematic review and meta-analysis aimed to estimate the prevalence of voice disorders and related factors in university professors.

METHODS

In this systematic review, Medline, Scopus, Embase, Cochrane Library, Eric, ProQuest, Magiran, Scientific Information Database and IranDoc databases were searched. The search was limited to January 1990 and May 2020. The inclusion criteria were reports of the prevalence of voice disorders in university professors and original studies in English and Persian languages. The exclusion criteria were studies assessing voice disorders in school or music teachers; the full-text not available; and case studies, conference papers, and review studies. All eligible studies were selected and critically appraised using the Joanna Briggs Institute checklist. Finally, a meta-analysis was performed using STATA 16.0 statistical software.

RESULTS

The preliminary search yielded 1251 articles and 18 of which met the eligibility criteria. The overall prevalence of voice disorders in university professors was 41% (Pooled prevalence: 0.41, 95% CI: 0.34-0.49, P-value < 0.001). Country-based analysis showed that the highest prevalence of voice disorders among university professors was in Iran 69% (Pooled prevalence: 0.69, 95% CI: 0.62-0.76) and the lowest was in China 20% (Pooled prevalence: 0.20, CI: 0.14-0.27). Among 4037 university professors, caffeine consumption was the most frequent related factor (87%) and dry throat was the most frequent symptom (46%).

CONCLUSIONS

The present study yielded about 41% of the professors had voice disorders. Among the influential habits, caffeine consumption and among the symptoms, dry throat were very common in university professors. Due to the cross-sectional nature of our studies, we were not able to perform further analyses on the risk factors for voice disorders. Therefore, more longitudinal surveys are needed for reaching a more reliable and deep view into the development of voice disorders.

Vocal Sensory Symptoms, Vocal Fatigue and Vocal Habits in University Professors

PURPOSE

To identify vocal sensory symptoms, vocal fatigue, and vocal habits, compare and correlate them in university professors who have passed or failed a screening related to vocal symptoms.

STUDY DESIGN

Cross-sectional, quantitative, non-randomized study.

METHODS

The professors responded to four vocal self-assessment Patient-Reported Outcome Measures (PROM): the Brazilian validated version of the Voice Symptom Scale (VoiSS), the Brazilian version (not validated) of the Vocal Tract Discomfort Scale, the Brazilian validated version of the Vocal Fatigue Index and the validated version of the Vocal Health and Hygiene Questionnaire ("Questionário de Saúde e Higiene Vocal - QSHV," in Portuguese). After the PROM's application, the professors were divided into two groups from the VoiSS cutoff point. G1 was formed by professors who passed in the VoiSS screening, that is, those who achieved a total score below 16 points, and G2, formed by professors who failed, that is, those who achieved 16 points or more. The Mann-Whitney U test was used to compare the values of PROM between G1 and G2 groups, and the Spearman correlation test was used to correlate the PROM with one another.

RESULTS

One hundred and two professors filled in the PROM, 23 from G1 and 79 from G2. Results showed that the total scores of all PROM used exceeded their cutoff scores in the overall sample. It was also observed that professors who failed the screening through VoiSS presented higher scores of vocal sensory symptoms and vocal fatigue, however no differences regarding knowledge on vocal health and hygiene were noticed. In G1, the protocols showed significant correlations ranging from moderate to very strong with one another, but the QSHV showed no correlation with any of the instruments. In G2, the protocols presented weak to strong correlation ratios, and the QSHV reported correlation with all protocols, except for Vocal Fatigue Index.

CONCLUSION

University professors presented scores higher than the cut-off point suggested by the literature for vocal sensory symptoms, vocal fatigue and knowledge on vocal health and hygiene. Professors who failed the screening performed through VoiSS achieved higher scores of vocal discomfort and fatigue, but also presented a high score for knowledge on vocal hygiene. There was a moderate to very strong statistical correlation among the vocal symptoms, discomforts, vocal fatigue and vocal habits in professors who failed the screening through VoiSS, however among those professors who passed the screening, knowledge on vocal health and hygiene showed no correlation with the PROM analyzed.

Replication-based regularization approaches to diagnose Reinke's edema by using voice recordings

Reinke's edema is one of the most prevalent laryngeal pathologies. Its detection can be addressed by using computer-aided diagnosis systems based on features extracted from speech recordings. When extracting acoustic features from different voice recordings of a particular subject at a concrete moment, imperfections in technology and the very biological variability result in values that are close, but they are not identical. This suggests that the within-subject variability must be properly addressed in the statistical methodology. Regularization-based regression approaches can be used to reduce the classification errors by favoring the best predictors and penalizing the worst ones. Three replication-based regularization approaches for variable selection and classification have been specifically designed and implemented to take into account the underlying within-subject variability. In order to illustrate the applicability of these approaches, an experiment has been specifically conducted to discriminate Reinke's edema patients (30 subjects) from healthy people (30 subjects) in a hospital environment. The features have been extracted from four phonations of the sustained vowel /a/ recorded for each subject, leading to a database that has fed the proposed machine learning approaches. The proposed replication-based approaches have been proved to be reliable in terms of selected features and predictive ability, leading to a stable accuracy rate of 0.89 under a cross-validation framework. Also, a comparison with traditional independence-based regularization methods reports a great variability of the latter in terms of selected features and accuracy metrics. Therefore, the proposed approaches contribute to fill a gap in the scientific literature on statistical approaches considering within-subject variability and can be used to build a robust expert system.

How can technology assist occupational voice users?

SUMMARY

The voice is an important tool for people who use it daily in their occupations. However, what technological options are available to such individuals to allow them to monitor or take care of their voices?

OBJECTIVE

The purpose of this study is to answer two research questions: (1) What technologies exist to monitor or take care of the voice in occupational voice users? (2) What is the technology readiness level (TRL) of the technologies used to monitor or take care of the voice in occupational voice users?

DATA SOURCES

Embase, IEEE, Medline, Proquest, PubMed, Scopus, and Web of Science.

METHODS

A systematic literature review was conducted. Articles that reported results regarding technologies (hardware, software, or mobile apps) that were used to monitor or take care of the voice in occupational voice users were included.

RESULTS

After reviewing 4581 abstracts, 10 full text studies were included in the literature review. The technologies found include 30% hardware, 30% hardware plus software, and 50% mobile apps, with an overall TRL mean of 5.3 (SD = 2.3).

CONCLUSION

Further research is necessary for higher validity in the studies and to increase the readiness in the development of current technologies to offer more options for this population.Implications for RehabilitationThe evidence for the impact of the use of the technologies for occupational voice users is still lowThere is emerging evidence that mobile apps and artificial intelligence algorithms can be used to investigate vocal disorders or potential risks in occupational voice usersMore research is required to increase the readiness developmental stage of current technologies for occupational voice users.

Occupational voice is a work in progress: active risk management, habilitation and rehabilitation

The current article reviews recent literature examining occupational voice use and occupational voice disorders (January 2018–July 2019).