• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.25.2022.tde-09032023-084459
Document
Author
Full name
Gabriela Aparecida Prearo
Institute/School/College
Knowledge Area
Date of Defense
Published
Bauru, 2022
Supervisor
Committee
Dutka, Jeniffer de Cassia Rillo (President)
Fabron, Eliana Maria Gradim
Ferreira, Gabriela Zuin
Teles, Lidia Cristina da Silva
Title in Portuguese
Espectro médio de longo termo da fala de indivíduos com fissura labiopalatina com diferentes graus de hipernasalidade
Keywords in Portuguese
Acústica da fala
Análise espectral
Distúrbios da fala
Fala
Fissura palatina
Insuficiência velofaríngea
Abstract in Portuguese
Objetivos: 1-Descrever e comparar as características do Espectro Médio de Longo Termo (EMLT) de amostras de fala sem e com hipernasalidade, nas vozes masculina e feminina de crianças e adultos; 2- Desenvolver um indicador acústico do grau da hipernasalidade de fala a partir da análise espectral. Métodos: Um total de 192 gravações de fala foram selecionadas para o estudo e foram agrupadas de acordo com a história de fissura e o grau de hipernasalidade: Grupo-SemFi: SEM fissuras e SEM hipernasalidade (n=40); Grupo- ComFi-Ausente: COM fissura e SEM hipernasalidade (n=55); Grupo-ComFi-Leve: COM fissura e COM hipernasalidade de grau leve (n=38); Grupo-ComFi-Moderada: COM fissura e COM hipernasalidade de grau moderada (n=59). As 192 gravações foram editadas criando-se duas amostras nos contextos de interesse do estudo: amostras representativas do contexto oral (n=192) e amostras representativas do contexto nasal (n=192). A análise acústica por meio do EMLT foi realizada para cada uma das 384 amostras usando-se o software Praat. Os valores de amplitudes do EMLT foram extraídos com intervalos de 0,1kHz, a frequência máxima analisada foi de 4kHz e os valores estabelecidos foram convertidos em Pascal. Valores médios foram estabelecidos para os 4 grupos estudados (SemFi, ComFi-Ausente, ComFi-Leve e ComFi-Moderada), nos dois tipos de vozes (masculina e feminina) e nas duas faixas etárias (crianças e adultos). A análise discriminante linear (ADL) foi realizada buscando-se um indicador acústico para os diferentes graus de hipernasalidade. Resultados: Os resultados do EMLT demonstraram diferenças significativas entre as vozes masculina e feminina, entre as vozes de adultos e crianças e entre os graus de hipernasalidade avaliadas (SemFi, ComFi- Ausente, ComFi-Leve e ComFi-Moderada). A ADL realizada com os resultados da análise pelo EMLT revelou fórmulas que classificaram corretamente 75% dos diferentes graus de hipernasalidade. Conclusão: Os diferentes tipos de amostras estudadas resultaram em espectros diferentes possibilitando
Title in English
Long-term average spectra at the speech of individuals with cleft lip and palate with different degrees of hypernasality
Keywords in English
Cleft palate
Spectrum analysis
Speech
Speech acoustics
Speech disorders
Velopharyngeal insufficiency
Abstract in English
Objectives: 1-To describe and compare the characteristics of the Long-term Average Spectra (LTAS) of speech samples without and with hypernasality, in male and female voices of children and adults; 2- Develop an acoustic indicator of the degree of speech hypernasality from spectral analysis. Methods: A total of 192 speech recordings were selected for the study and were grouped according to cleft history and degree of hypernasality: Group SemFi: WITHOUT clefts and WITHOUT hypernasality (n=40); Group ComFi-Ausente: WITH cleft and WITHOUT hypernasality (n=55); Group ComFi-Leve: WITH cleft and WITH mild hypernasality (n=38); Group ComFi- Moderada: WITH cleft and WITH moderate degree hypernasality (n=59). The 192 recordings were edited creating two samples in the contexts of interest to the study: representative samples of the oral context (n=192) and representative samples of the nasal context (n=192). Acoustic analysis using EMLT was performed for each of the 384 samples using Praat software. The EMLT amplitude values were extracted with intervals of 0.1kHz, the maximum analyzed frequency was 4kHz and the established values were converted into Pascal. Mean values were established for the 4 groups studied (SemFi, ComFi-Ausente, ComFi-Leve and, ComFi-Moderada), in both types of voices (male and female) and in both age groups (children and adults). Linear discriminant analysis (ADL) was performed seeking an acoustic indicator for the different degrees of hypernasality. Results: The EMLT results showed significant differences between male and female voices, between adult and children's voices and between the degrees of hypernasality evaluated (SemFi, ComFi- Ausente, ComFi-Leve and ComFi-Moderada). The ADL performed with the results of the analysis by the EMLT revealed formulas that correctly classified 75% of the different degrees of hypernasality. Conclusion: The different types of samples studied resulted in different spectra, making it possible to establish the preliminary characteristics of the EMLT of speech without and with hypernasality, in the male and female voices of children and adults. The ADL resulted in formulas that correctly classified most of the different degrees of hypernasality represented in the studied samples. Future studies are needed to investigate the potential of the created formulas.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2023-03-09
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.