Modelagem matemática-computacional das pregas vocais aplicada à síntese de vozes alteradas

• Ano de defesa: 2021
• Autor(a) principal: Santos, Jônatas Cruz
• Orientador(a): Montalvão Filho, Jugurta Rosa
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Programa de Pós-Graduação: Pós-Graduação em Engenharia Elétrica
• Palavras-chave em Português: Cordas vocais; Modelos matemáticos; Vozes; Distúrbios da voz; Pregas vocais; Modelagem matemática-computacional; Alterações vocais
• Palavras-chave em Inglês: Vocal folds; Computational modelling; Vocal disturbance
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: http://ri.ufs.br/jspui/handle/riufs/16172

The use of mathematical-computational models of phonation has been developed since the 60’s, being improved throughout the years. This models are very useful by allowing the reproduction of vocal dynamics and of vocal folds behavior during voice and speech emission, being able to help the advances of vocal mechanisms comprehension and its activities related to voice production, also allowing the generation of synthetic voice signals in different conditions. In this work, the anatomical and physiological elements that take part in phonation and the main models for the vocal folds and vocal tract are presented. Based on that, an adaptation of literature models is proposed, allowing the representation of voices with a vocal pathology (vocal polyp). Based on the synthetic samples generated by this model, analytical conjectures are performed, show a high potential for improving polyp detectors at human voice samples performance. Also, a synthesizer implementation is done, based on a well established model on literature, able to generate synthetic voice samples, being controlled by muscular and aerodynamic parameters of voice production, also allowing the adjust of voice disturbance conditions. That synthesizer might be useful, being applied, for example, at the vocal dynamics study and helping vocal evaluation by means of analysis-by-synthesis methods.