Influência da experiência do ouvinte e da tarefa de fala na avaliação perceptivo-auditiva da qualidade vocal

• Autor(a) principal: Alves, Jônatas do Nascimento
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFPB
• Texto Completo: https://repositorio.ufpb.br/jspui/handle/123456789/18733
• topic: Análise perceptivo auditiva; Tarefa de fala; Experiência; Confiabilidade; Acurácia; Auditory-perceptual analysis; Speech task; Experience; Reliability; CNPQ::LINGUISTICA, LETRAS E ARTES::LINGUISTICA

Introduction: Voice quality is a perceptual phenomenon by nature. In the routine of the speech-language pathology clinic, auditory-perceptual analysis of voice, despite its subjectivity, remains the reference standard in voice evaluation. Several factors may influence the auditory-perceptual analysis of voice, for example, judge experience, type of speech task evaluated, and measurement scale used. However, many doubts remain about the roles of these factors in the auditory-perceptual analysis of voice quality. Objective: To assess whether listener experience, the type of speech task, and the auditory-perceptual analysis of vocal deviation intensity are associated with the predominant voice quality. Method: In total, 22 judges, divided into four groups, participated in the study: speech-language pathologists specialized in voice, general speech-language pathologists, graduate students with experience in auditory-perceptual analysis and graduate students without experience. Two reference judges also participated. The subjects rated the predominant voice quality and the vocal deviation intensity of a sample of 44 voices. Various speech tasks were used, namely, sustained vowels /a/ and /Ɛ/, Consensus Auditory-Perceptual Evaluation of Voice (CAPE-V) sentences, number counting, and running speech. The speech tasks were presented in two modalities. In the first, the judges rated the speech tasks separately, that is, all vowels /a/, all vowels /Ɛ/, and so forth. Then, the speech tasks were presented to the judges interconnectedly. Two measurement scales were used, the visual analog scale and the numerical scale. Results: The general speech-language pathologists and by trained graduate students showed the best interrater reliability indices. The CAPE-V sentences task showed the best reliability among all tasks. No significant differences in the reliability of the auditory-perceptual analysis were found between the vowel /a/ and vowel /ɛ/ tasks. In general, the visual analog scale had higher interrater reliability than the numerical scale. The specialist speech therapists showed the best accuracy rates in the auditory perceptual analysis of vocal quality. The vowel speech task /Ɛ/ was the one with the highest accuracy values in all groups in relation to the reference judge. Conclusion: Listener experience and speech task affected the reliability also the accuracy of the auditory-perceptual analysis of voice quality