Applied Chaos Level Test for Validation of Signal Conditions Underlying Optimal Performance of Voice Classification Methods.

Purpose: The purpose of this study is to introduce a chaos level test to evaluate linear and nonlinear voice type classification method performances under varying signal chaos conditions without subjective impression. Study Design: Voice signals were constructed with differing degrees of noise to model signal chaos. Within each noise power, 100 Monte Carlo experiments were applied to analyze the output of jitter, shimmer, correlation dimension, and spectrum convergence ratio. The computational output of the 4 classifiers was then plotted against signal chaos level to investigate the performance of these acoustic analysis methods under varying degrees of signal chaos. Method: A diffusive behavior detection-based chaos level test was used to investigate the performances of different voice classification methods. Voice signals were constructed by varying the signal-to-noise ratio to establish differing signal chaos conditions. Results: Chaos level increased sigmoidally with increasing noise power. Jitter and shimmer performed optimally when the chaos level was less than or equal to 0.01, whereas correlation dimension was capable of analyzing signals with chaos levels of less than or equal to 0.0179. Spectrum convergence ratio demonstrated proficiency in analyzing voice signals with all chaos levels investigated in this study. Conclusion: The results of this study corroborate the performance relationships observed in previous studies and, therefore, demonstrate the validity of the validation test method. The presented chaos level validation test could be broadly utilized to evaluate acoustic analysis methods and establish the most appropriate methodology for objective voice analysis in clinical practice.

Investigation of the Effects of Speech Signal Length on Vocal Disorder Sorting Done Via Dynamic Pattern Modeling.

OBJECTIVES: Development of a noninvasive method for separating different vocal fold diseases is an important issue concerning vocal analysis. Due to the time variations along a pathologic vocal signal, application of dynamic pattern modeling tools is expected to help in the detection of defects that occur in the speech production mechanism. MATERIALS AND METHODS: In the present study, the hidden Markov model, which is a state space model, is employed to sort some of the vocal diseases. Moreover, this research mainly investigates the effects of the processed vocal signal lengths on the mentioned sorting task. To this end, the signal lengths of 1, 3, and 5 seconds of different disorders are used. RESULTS: The experimental results show that some pathologic conditions in vocal folds such as cyst, false vocal cord, and mass are more evident in continued voice production, and the recognition accuracies gained via dynamic modeling of pathologic voice signals with more lengths are considerably improved.

Analysis and Classification of Voice Pathologies Using Glottal Signal Parameters.

The classification of voice diseases has many applications in health, in diseases treatment, and in the design of new medical equipment for helping doctors in diagnosing pathologies related to the voice. This work uses the parameters of the glottal signal to help the identification of two types of voice disorders related to the pathologies of the vocal folds: nodule and unilateral paralysis. The parameters of the glottal signal are obtained through a known inverse filtering method, and they are used as inputs to an Artificial Neural Network, a Support Vector Machine, and also to a Hidden Markov Model, to obtain the classification, and to compare the results, of the voice signals into three different groups: speakers with nodule in the vocal folds; speakers with unilateral paralysis of the vocal folds; and speakers with normal voices, that is, without nodule or unilateral paralysis present in the vocal folds. The database is composed of 248 voice recordings (signals of vowels production) containing samples corresponding to the three groups mentioned. In this study, a larger database was used for the classification when compared with similar studies, and its classification rate is superior to other studies, reaching 97.2%.

Aerodynamic Outcomes of Four Common Voice Disorders: Moving Toward Disorder-Specific Assessment.

OBJECTIVES: The two goals of the present study were to (1) determine the ability of commonly used aerodynamic voice measures to capture change as a function of known interventions and (2) determine if certain aerodynamic measures demonstrate better responsiveness to change in specific disorder types than others. STUDY DESIGN: This is a retrospective, longitudinal, single-blinded, cross-sectional study. METHODS: Patients (n = 70) with a single voice disorder diagnosis of benign vocal fold lesions (lesions), unilateral vocal fold paralysis (UVFP), primary muscle tension dysphonia (MTD-1), or vocal fold atrophy (atrophy) underwent baseline testing, a single intervention (phonosurgery or voice therapy), and follow-up testing. Common aerodynamic measurements were completed in repeated syllables and an all-voiced sentence. RESULTS: Statistically significant improvements were observed for two outcome measures, average airflow in syllables, and average airflow in the all-voiced sentence. Patients with lesions, UVFP, and MTD-1 improved in average airflow in the all-voiced sentence. Patients with UVFP also improved in airflow in syllables. CONCLUSIONS: Average airflow in the all-voiced sentence changed as a function of treatment for the lesion, MTD-1, and UVFP groups, demonstrating a disorder-specific pattern. Laryngeal airway resistance, and estimates of average subglottal pressure did not show significant change. Average airflow in the all-voiced sentence measurements is recommended as a routine voice measure, and further investigation of other aerodynamic measures' sensitivity to change is warranted.

[Assessment of voice in patients with Reinke's oedemas treated with laser microsurgery]. / Ocena glosu u chorych z obrzekami Reinkego leczonych operacyjnie z uzyciem lasera CO2.

This investigation was undertaken for perceptual and acoustic analysis of voice in patients with Reinke's oedemas treated with laser microsurgery. Material of the study included 45 patients in the age of 23 to 65 presenting to Foniatric Clinic with dysphonia in Reinke's oedemas. Patients were divided into 3 groups according to Yonakawa's classification of Reinke's oedemas. All patients were treated with CO2 laser microsurgery. Voice assessment was made subiectively with GRBAS scale and objectively using multidimensional voice parameters analysis (MDVP). The examination was undertaken preoperatively and 4 weeks after operation. The results of our study show relationship between quality of voice after surgery and the stage of evolution of Reinke's oedemas according to the applied classification.

[Does the subjective assessment of roughness correlate with a computer-aided measurement of irregularity index?]. / Objektiv gemessene Stimmlippenschwingungsirregularität vs. subjektiver Eindruck der Rauigkeit.

For routine clinical purposes dysphonic voices are assessed using the GRBAS scale or analogues. An objective measurement often includes jitter measurements. Here we raised the question of whether roughness estimates correlate with a computer-aided measurement of the vocal fold cycle irregularity in 78 patients who read a standard text ("Nordwind und Sonne"). The samples were evaluated by 19 speech and voice therapy students according to the degree of roughness. The irregularity index was calculated according to the procedures suggested by Fourcin. Data were subjected to correlation analysis. A significant correlation was found between the irregularity index and the subjective roughness scaling. These data indicate that objective measurements of vibratory cycle irregularity during text reading can be used for clinical purposes in addition to subjective roughness scaling.

Automatic assessment of voice quality according to the GRBAS scale.

Nowadays, the most extended techniques to measure the voice quality are based on perceptual evaluation by well trained professionals. The GRBAS scale is a widely used method for perceptual evaluation of voice quality. The GRBAS scale is widely used in Japan and there is increasing interest in both Europe and the United States. However, this technique needs well-trained experts, and is based on the evaluator's expertise, depending a lot on his own psycho-physical state. Furthermore, a great variability in the assessments performed from one evaluator to another is observed. Therefore, an objective method to provide such measurement of voice quality would be very valuable. In this paper, the automatic assessment of voice quality is addressed by means of short-term Mel cepstral parameters (MFCC), and learning vector quantization (LVQ) in a pattern recognition stage. Results show that this approach provides acceptable results for this purpose, with accuracy around 65% at the best.

Inter-rater reliability of an oral assessment guide for elderly patients residing in a rehabilitation ward.

The aim of this study was to test the inter-rater reliability of a revised oral assessment guide (ROAG) for patients residing in a geriatric rehabilitation ward. A consecutive sample of 140 patients was recruited for the study. Oral assessments were performed for 133 newly admitted patients by one registered nurse (RN) during a period of six months. A dental hygienist (DH) carried out 103 oral assessments during the same half-year. For 66 patients, the RN and the DH performed independent assessments. There was an agreement between the RN and the DH in the majority of the independent assessments, except for tongue and teeth/dentures. The percentage agreement exceeded 80 percent. Inter-rater agreement measured by Cohen's Kappa coefficient ranged from moderate to very good and percentage agreement had a range of 58 to 91 percent. The agreement was highest in assessment of voice and swallowing (91%). Assessments of teeth and dentures seemed to be most difficult for the RN to evaluate. ROAG was found to be a clinically useful assessment tool. Additional education and training is needed to improve the reliability of the oral assessments and should include continuous support from a dental hygienist as well as a pictorial manual on how to use the ROAG.

Assessment of two objective voice function indices.

In the care of patients with voice disorders, physicians, speech pathologists, and other health care professionals routinely make diagnoses, recommend treatment, and evaluate outcomes. Although objective and subjective measures exist, unfortunately, there is no widely accepted, valid method for classifying voice disorders and assessing outcome after voice treatment. In the present research, the relationship between two previously created multivariate objective voice function indices, the weighted odds ratio index and the multivariate logistic regression index, and subjective assessment of voice function was evaluated. Twenty-three adult patients presenting to a speech science laboratory for evaluation of voice disorders were studied in this prospective observational study together with 12 normal volunteers as controls. Vocal function was measured on 14 different parameters with a protocol that included a multichannel input for simultaneous assessment of acoustic and physiological parameters. Each patient was recorded reading the standard passage "The North Wind and the Sun," and recordings were then evaluated by the GRBAS scale. Overall, there was a statistically significant relationship between the weighted odds ratio index and multivariate logistic regression index and mean GRBAS scores. This research demonstrates that the voice function values calculated from two different multivariate objective voice function indices are significantly associated with subjective voice assessments. These multivariate objective voice indices may be appropriate for use in clinical trials and outcomes research on treatment effectiveness for voice disorders.