Treinamento auditivo com estímulos vocais âncoras sintetizados: efeito na concordância dos avaliadores

Abstract

Objective: to analyze the results from the perceptual training, with synthesized auditory anchors, in the intrarater agreement of the perceptual rating of roughness and breathiness. Method: The research was approved by the Research Ethics Committee (37872314.2.0000.5149). This is an experimental study consisting on four perceptual training sessions with human voices and synthesized anchors. The sample consisted on twenty Speech Language Pathologist students who had previous contact with auditory-perceptual assessment, aged 21 to 37 years old, with an average of 24 years old, three men and seventeen women. The evaluators participated on four training sessions with an interval of seven days between them. Each training consisted of three tasks: 1) Pre-training activity: judgment of 20 normal and dysphonic voices, where the participants evaluated the parameters of roughness and breathiness and the degree of vocal deviation (0 - neutral, 1 - mild, 2 - moderate, 3 - intense); 2) Training activity: four synthesized auditory anchors for roughness (R) and four synthesized auditory anchors for breathiness (B) were presented with the general degree of vocal deviation ranging from zero to three. Participants heard four natural voice stimuli and an anchor stimulus, and were instructed to pair the natural voice that most closely resembled the synthesized anchor stimulus. Participants performed the training with 20 natural voices; and 3) Post-training activity: the 20 voices from the pre-training activity were randomized and the individuals judged the same voices without prior knowledge that the voices were repeated. Statistical analysis of the data was performed using the AC2 test to assess intrarater agreement and the Friedman non-parametric test for comparison between training sessions. The software used was R (version 3.5.1) to perform the AC2 test and Minitab® 19 for the Friedman test. A significance level of 5% was considered. Results: In the analysis of the intrarater agreement for the perceptual-auditory parameter of roughness, the results varied between 79%, 85%, 85% and 86% between the first and the fourth auditory training session, respectively, with improved intrarater agreement from the fourth session (p = 0.005). For the analysis of the intrarater agreement on the breathiness, the results were 88%, 90%, 90% and 92% from the first to the fourth auditory training session, respectively. In the fourth training session, the judges showed a greater agreement (p = 0.036). Conclusion: The perceptual-auditory parameter of breathiness showed a higher indicator AC2 than for roughness, suggesting that it was more consistent. As in the case of roughness, breathiness also showed improvement in intrarater agreement from the fourth session of auditory training.