Uso do programa MDVP em diferentes contextos: revisão de literatura

HEAR set: A ligHtwEight acoustic paRameters set to assess mental health from voice analysis

BACKGROUND

Voice analysis has significant potential in aiding healthcare professionals with detecting, diagnosing, and personalising treatment. It represents an objective and non-intrusive tool for supporting the detection and monitoring of specific pathologies. By calculating various acoustic features, voice analysis extracts valuable information to assess voice quality. The choice of these parameters is crucial for an accurate assessment.

METHOD

In this paper, we propose a lightweight acoustic parameter set, named HEAR, able to evaluate voice quality to assess mental health. In detail, this consists of jitter, spectral centroid, Mel-frequency cepstral coefficients, and their derivates. The choice of parameters for the proposed set was influenced by the explainable significance of each acoustic parameter in the voice production process.

RESULTS

The reliability of the proposed acoustic set to detect the early symptoms of mental disorders was evaluated in an experimental phase. Voices of subjects suffering from different mental pathologies, selected from available databases, were analysed. The performance obtained from the HEAR features was compared with that obtained by analysing features selected from toolkits widely used in the literature, as with those obtained using learned procedures. The best performance in terms of MAE and RMSE was achieved for the detection of depression (5.32 and 6.24 respectively). For the detection of psychogenic dysphonia and anxiety, the highest accuracy rates were about 75 % and 97 %, respectively.

CONCLUSIONS

The comparative evaluation was carried out to assess the performance of the proposed approach, demonstrating a reliable capability to highlight affective physiological alterations of voice quality due to the considered mental disorders.

Acoustic Characteristics of the Voice for Brazilian Portuguese Speakers Across the Life Span

INTRODUCTION

Vocal changes occur across the life span and can be reflected in acoustic measurements.

OBJECTIVE

To investigate the characteristics of voice production of the Brazilian Portuguese speakers across the life span based on acoustic measures of Fundamental Frequency (f_o) and noise-to-harmonic ratio (NHR) and to verify the differences in these measures between men and women.

METHODS

A total of 526 recordings from Brazilian Portuguese speakers aged 5-93 years were included. Voices from these speakers were judged with normal vocal quality for their age using the G parameter of the GRBAS scale. The recordings were divided into 12 age groups (5-7 years; 8-9; 10-11; 12; 13-15; 16-18; 19-29; 30-39; 40-49; 50-59; 60-69 and, 70-93 years old). Acoustic analysis was conducted, extracting the parameters f_o and NHR through Multi-Dimensional Voice Programsoftware.

RESULTS

For women, there was a gradual decrease in f_o from childhood to older age. Older women (60-93 years old) showed a lower f_o than age groups up to 19-29 years (P< 0.00). For men, there was a decrease in f_o up to the age group of 13-15 years (P< 0.00) and, then, it remained stable. Differences between sexes occurred from 12 years old, with higher f_o values for women than men (P< 0.00). The NHR parameter remained stable across the life span for women while higher values for older subjects were found for men (P< 0.04). Regarding sex, men showed a higher NHR value than women (P< 0.002).

CONCLUSION

Vocal changes occur across the life span and are reflected in the acoustic measure of f_o for men and women. The beginning of vocal changes occurs from 12 years old with differences between sexes. The NHR measure was sensitive to indicate changes over a lifetime for men, with higher values for older subjects.

Worldwide Healthy Adult Voice Baseline Parameters: A Comprehensive Review

The voice results in acoustic signals analyzed and synthetized at first for telecommunication matters, and more recently investigated for medical purposes. In particular, voice signal characteristics can evidence individual health conditions useful for screening, diagnostic and remote monitoring aims. Within this frame, the knowledge of baseline features of healthy voice is mandatory, in order to balance a comparison with their unhealthy counterpart. However, the baseline features of the human voice depend on gender, age-range and ethnicity and, as far as we know, no work reports as those features spread worldwide. This paper intends to cover this lack. Our database research yielded 179 relevant published studies, retrieved using digital libraries of IEEE Xplore, Scopus, Web of Science, Iop Science, Taylor and Francis Online, and Scitepress. These relevant studies report different features, among which here we consider the most investigated ones, within the most investigated age-range. In particular, the features are the fundamental frequency, the jitter, the shimmer, the harmonic-to-noise ratio, and the cepstral peak prominence, the most investigated age-range is within 20-40 years and, related to the ethnicity, 20 countries are considered.

Evidence of Internal Consistency in the Spectrographic Analysis Protocol

OBJECTIVE

To verify the validity in the internal consistency in the spectrographic analysis protocol (SAP).

MATERIAL AND METHODS

Thirty-nine students of the Speech-Language Pathology graduate program and 38 speech-language pathologists, specialized in voice, participated in the study. The participants made visual inspections of 10 spectrograms and marked the items of the SAP. For analysis of the internal consistency in the SAP, the exploratory factor analysis (EFA) and confirmatory factor analysis were performed.

RESULTS

Most items showed corrected item-total correlation above 0.3, indicating that the items have a good relationship with each other and with the SAP as a whole. Six items presented values below the average, suggesting the exclusion of these from the construct. However, three of these were maintained because they were judged as important parameters in clinical practice, requiring the training of judges when using the SAP to properly understand the items. The EFA regrouped the previous domains of the SAP into three factors. All items presented a factor load above 0.4, suggesting the retention of all, except for the items previously indicated, for exclusion. The confirmatory factor analysis corroborated with the EFA and its indexes.

CONCLUSION

The SAP has good internal consistency. All items have a good degree of relationship with each other and contribute positively to the protocol as a whole. The final version of the SAP, at this stage, has 15 items (from the 25 items of the initial SAP version), distributed among three domains.

Instrumental Acoustic Voice Characteristics in Adults with Type 2 Diabetes

OBJECTIVE

The objective of this study was to investigate if there are differences in acoustic parameters between diabetic patients and normal controls.

METHODS

A prospective cross-sectional study was performed in 83 diabetic patients and 70 healthy controls. Voice parameters including fundamental frequency (F0), jitter, shimmer, amplitude perturbation quotient, noise-to-harmonic ratio, smoothed amplitude perturbation quotient, and relative average perturbation were analyzed using Computerized Speech Lab with the Multi-Dimensional Voice Program.

RESULTS

F0 in female diabetic patients was significantly lower than controls (222.23 ± 27.89 Hz versus 241.08 ± 28.21 Hz, P< 0.01). In female diabetic subgroups with disease duration more than 10 years, poor glycemic control, or neuropathy, the F0 was still significantly lower. Multivariate analysis showed that F0 was significantly associated with diabetes after controlled for age, body mass index, presence of hypertension, and dyslipidemia. (P= 0.022). However, F0 was not able to predict the presence of diabetes as shown by logistic regression analysis (P= 0.243).

CONCLUSIONS

Voice fundamental frequency is lower in females with diabetes. However, voice fundamental frequency cannot adequately predict the presence of diabetes.

Acoustic Measurements of the Glottic Source of Female Teachers With Dysphonia

OBJECTIVE

To verify the acoustic measurements of glottic source of dysphonic teachers of a medium-sized municipality of interior of the state.

METHODS

Retrospective, cross-sectional, and quantitative study, with a composite sample of 34 dysphonic teachers, of which 21 teachers without laryngeal affections and 13 with laryngeal affections, mean age 39.1 years old and 39.5 years old, respectively. Glottic source acoustic analysis was performed with the Multi-Dimensional Voice Program Advanced. The data were analyzed statistically to verify the significance of each acoustic measure between the groups (with laryngeal affection, without laryngeal affection, and total) and in relation to the normality proposed by the software.

RESULTS

In the three conditions (groups with and without affection and total) the means were statistically below normality in the measurements of maximum and minimum fundamental frequency. In the group without affection, frequency, and noise measurements presented above normality. In both groups, measurements of frequency, noise, and subharmonic segments were above normal, and number of voice breaks below normal.

CONCLUSION

Acoustic parameters outside the normal pattern showed an aperiodic vocal production, with presence of noise and instability in the vocal signal, in dysphonic teachers with or without alteration at the laryngeal level.

Acoustic characteristics of voice in different cycles of life: an integrative literature review

ABSTRACT Purpose: to carry out an integrative literature review about the acoustic characteristics of healthy voice production, from childhood to old age. Methods: a bibliographic survey was conducted on the databases PubMed, SciELO, MEDLINE and LILACS, covering the last 10 years. Nineteen studies were found, meeting the proposed criteria, on acoustic measurements: F0 (fundamental frequency), jitter, shimmer and/or noise measurements, in males and females, with normal voices in their different stages of life. Results: the analysis showed that F0 is the most changing acoustic parameter as people grow up and grow old. Its values present gradual fall from childhood to old age in the female population, whereas among men such decrease lasts until adulthood. Jitter, shimmer and noise remain stable throughout childhood and adulthood, while shimmer and noise measurements increase in old age. In the literature, there is no consensus regarding increase of jitter measurements in the elderly. Conclusion: from childhood to old age, in both genders, vocal changes take place which are reflected, especially by F0. There is a scarcity of information on acoustics related to specific populations with ample age range, using the same methodology. The information in this study may guide future investigations aiming to understand natural changes occurring in the human voice, in addition to guiding in the clinical practice.

An Investigation of Multidimensional Voice Program Parameters in Three Different Databases for Voice Pathology Detection and Classification

BACKGROUND AND OBJECTIVE

Automatic voice-pathology detection and classification systems may help clinicians to detect the existence of any voice pathologies and the type of pathology from which patients suffer in the early stages. The main aim of this paper is to investigate Multidimensional Voice Program (MDVP) parameters to automatically detect and classify the voice pathologies in multiple databases, and then to find out which parameters performed well in these two processes.

MATERIALS AND METHODS

Samples of the sustained vowel /a/ of normal and pathological voices were extracted from three different databases, which have three voice pathologies in common. The selected databases in this study represent three distinct languages: (1) the Arabic voice pathology database; (2) the Massachusetts Eye and Ear Infirmary database (English database); and (3) the Saarbruecken Voice Database (German database). A computerized speech lab program was used to extract MDVP parameters as features, and an acoustical analysis was performed. The Fisher discrimination ratio was applied to rank the parameters. A t test was performed to highlight any significant differences in the means of the normal and pathological samples.

RESULTS

The experimental results demonstrate a clear difference in the performance of the MDVP parameters using these databases. The highly ranked parameters also differed from one database to another. The best accuracies were obtained by using the three highest ranked MDVP parameters arranged according to the Fisher discrimination ratio: these accuracies were 99.68%, 88.21%, and 72.53% for the Saarbruecken Voice Database, the Massachusetts Eye and Ear Infirmary database, and the Arabic voice pathology database, respectively.