Análise do efeito imediato do Shaker® em indivíduos com e sem queixa vocal

Abstract

ABSTRACT Introduction: semi-occluded vocal tract exercises (SOVTE) are based on partial occlusion of the vocal tract during vocalization, originally used by singers and voice professionals as a way to increase performance and vocal quality. SOVTE are widely accepted among speech therapists, as they help to release tension and reduce phonatory effort. Recently, a new SOVTE technique was described, called voiced high-frequency oral oscillation, performed using a device called Shaker®. During the execution of the exercises, there is partial occlusion in the anterior region of the vocal tract, which favors a more efficient and economical vocal production due to the retroflex resonance. Purpose: to analyze auditory-perceptual and acoustic vocal changes after performing a high-frequency oral oscillation exercise, using the Shaker® device associated with vocal emission for three minutes. Methods: an intra subject comparative experimental study was carried out, with the approval of the Research Ethics Committee (Opinion 5,179,611). The study included 50 individuals, 25 males, aged between 19 and 42 years (mean age 24.8 and standard deviation of 4.8 years) and 25 females, aged between 19 and 56 years (mean age 29.3 and standard deviation of 9.3 years). The research was carried out at the Functional Health Observatory in Speech Therapy of the Faculty of Medicine of UFMG, in an acoustically treated room. University officials, students and volunteers in general were invited to participate in the survey. Participants fulfilled the Vocal Symptoms Scale, which aims to identify possible vocal symptoms. Each question is scored from 0 (never) to 4 (always). Individuals who reached a score equal to or greater than 16 were considered to have vocal complaints. To evaluate the effect of the Shaker® device associated with vocal emission for three minutes, firstly, a numerical visual scale (NVS) on the self-perception of vocal discomfort was applied at the moment before the first recording. In sequence, the participant's voice was recorded. The recordings were performed on a notebook computer with an AMD Ryzen 5 3500U processor, with a Dolby audio sound card. To capture the voice, a unidirectional Lesson® HD 74 microphone (cardioid) was used, duly connected to the computer, positioned laterally to the participant's mouth at a distance of approximately five centimeters. Participants were submitted, in an acoustically treated room, to voice recording during the emission of the prolonged vowel /ɛ/, in maximum phonation time, and counting from 1 to 10. Next, they blew the Shaker® mouthpiece, emitting the vowel /u/ for three minutes, and then the voice was recorded again and the numeric self-perception scale regarding vocal discomfort was completed. The vocal samples were submitted to acoustic analysis using the VoxMetria software, version 5.0, to obtain the harmonic-to-noise ratio, fundamental frequency (f0), noise, glottal to noise excitation (GNE), jitter and shimmer. The cepstral measurements Cepstral Peak Prominence (CPP) and Cepstral Prominence-Smoothed (CPPS) were extracted from the Praat software version 6.2. The recordings were also submitted to auditory perceptual analysis performed by three speech therapists, voice specialists with experience in auditory-perceptual voice analysis, independently. The audios were renamed to make the analysis blind and sent in random order to the speech therapists judges, so these speech therapists, when analyzing the pairs of voices, did not know if each analyzed voice was pre- or post-exercise. These classified the pairs of voices as “improved”, “no change” or “worse”. After the collections, four groups of participants were obtained, which were analyzed separately: women with vocal complaints, women without vocal complaints, men with vocal complaints and men without vocal complaints. Twenty percent of the sample was randomly replicated to verify the intra-evaluator agreement, in the auditory-perceptual evaluation, obtaining 100% agreement for the three evaluators. There was disagreement between the three judges in seven pairs of voices and, in these cases, a fourth evaluator, a voice specialist with more than 20 years of experience in the area, evaluated the voices to define the most concordant answer. Statistical analysis of the data was performed using the MINITAB statistical program, version 17. Distribution analysis of quantitative variables was performed using the Anderson-Darling test. Vocal self perception, auditory-perceptual assessment and CPP and CPPS cepstral measurements were compared between the pre and post-exercise moments. For such comparisons, the paired T-parametric or non-parametric Wilcoxon test was used. A 95% confidence level was considered for the analyses. Results: after the exercise there was a decrease in jitter (p=0.048) in the group of men with complaints and shimmer in men without complaints (p=0.042). The group of women with complaints showed an increase in the harmonic-to-noise ratio (p=0.011), fundamental frequency (p=0.011), CPP (p=0.002) and CPPS (p=0.014) values of the sustained vowel and a decrease in jitter (p=0.014), while women without vocal complaints showed an increase in GNE (p=0.013). All groups showed a significant decrease in vocal discomfort (p=0.009; p=0.022; p=0.002; p=0.009, respectively). A statistically significant difference was observed for the self-perception of vocal discomfort, with a reduction in values for both variables after the exercise. The group with the highest number of individuals who achieved improvement in the auditory perceptual assessment was the group of men with complaints, in which nine (69.23%) achieved improvement and four (30.77%) did not present vocal alterations. Conclusion: the exercise of voiced high frequency oral oscillation, using the Shaker® device, performed for three minutes, was capable of promoting acoustic changes and alleviating vocal discomfort in all analyzed groups, with the greatest changes observed in the group of women with vocal complaint.