Submitted by PPG Letras (letraspg@pucrs.br) on 2018-01-03T17:00:31Z
No. of bitstreams: 1
Tese com capa Marcio.pdf: 15597602 bytes, checksum: ee2807dd0f16d970b2c29da908ad2a99 (MD5) / Approved for entry into archive by Tatiana Lopes (tatiana.lopes@pucrs.br) on 2018-01-04T16:52:25Z (GMT) No. of bitstreams: 1
Tese com capa Marcio.pdf: 15597602 bytes, checksum: ee2807dd0f16d970b2c29da908ad2a99 (MD5) / Made available in DSpace on 2018-01-04T17:03:15Z (GMT). No. of bitstreams: 1
Tese com capa Marcio.pdf: 15597602 bytes, checksum: ee2807dd0f16d970b2c29da908ad2a99 (MD5)
Previous issue date: 2017-10-16 / The present research aimed to establish an indicator of variability of acoustic
parameters used as elements of confrontation in the expert examination of Voice
Comparison. The Variability Index proposed here considers the relationship between
intraindividual and interindividual variability and is intended to serve as a reference to
determine the advantage of one acoustic parameter over another among those used in
Voice Comparison. Besides being obtained from the values of the vowel formants,
traditional parameters in the Voice Comparison, the Variability Index was obtained from
the nasal consonant formants and from the measurements of vocalic acoustic space and
Euclidean Distances between vowels, with the aim to verify the applicability of these
parameters in Voice Comparison examination. The stimulus that served as the basis for
the study were tonic syllables pronounced in a carrier sentence and the data collection,
performed repeatedly in different days and shifts, and aimed at the control of variability,
favoring a minimum variability of the acoustic parameters. As for the method, twenty
subjects participated in the data collection and the individual average coefficients of
variation of the set and general average coefficients of variation of the set served as
referents of intraindividual and interindividual variability respectively. As the results of
variability, although were not achieved with nasal consonants, suggest that the Variability
Index allows us to establish the best parameters to be used in the Voice Comparison. The
Index allowed to order the vowels and also supported the proposal that the measurements
of vocal acoustic space and Euclidean Distances are possible parameters to be used in
Voice Comparison. Unlike Acoustic Phonetics and Articulatory, underlying the research,
Articulatory Phonology composes the work as an accessory theoretical contribution in
relation to the interpretation of the data of the present doctoral dissertation, but it is
understood that it is a valuable theory to explain the linguistic data usually found in forensic
practice. The Euclidean Distances between the vertices of the vocalic triangle - [a], [i] and
[u] - served to propose an analysis of how the computational implementation of the Task
Dynamics model, TADA, treats vowels in a continuously way and consonants as a
categorical form. / A presente pesquisa buscou estabelecer um indicador de variabilidade de
par?metros ac?sticos utilizados como elementos de confronto no exame pericial de
Compara??o de Locutores. O ?ndice de Variabilidade proposto por este estudo considera
a rela??o entre variabilidade intraindividual e interindividual e pretende-se que sirva como
refer?ncia para determinar a vantagem de um par?metro ac?stico sobre outro dentre
aqueles utilizados na Compara??o de Locutores. Al?m de ser obtido a partir dos valores
dos formantes voc?licos, par?metros tradicionais na Compara??o de Locutores, o ?ndice
de Variabilidade foi obtido com base nos formantes das consoantes nasais e a partir das
medidas de espa?o ac?stico voc?lico e dist?ncias euclidianas entre vogais, com o intuito
de verificar a aplicabilidade desses par?metros na Per?cia de Compara??o de Locutores.
Os est?mulos que serviram de base para o estudo foram s?labas t?nicas pronunciadas
em frase ve?culo e a coleta, realizada repetidamente em dias e turnos diferentes, e visou
o controle da variabilidade, favorecendo uma variabilidade m?nima dos par?metros
ac?sticos. Quanto ? metodologia, vinte indiv?duos participaram da coleta e os coeficientes
de varia??o individual m?dio do conjunto e geral do conjunto serviram como referentes
de variabilidade intraindividual e interindividual respectivamente. Os resultados de
variabilidade, apesar de n?o terem sido bem-sucedidos no que diz respeito ?s
consoantes nasais, sugerem que o ?ndice de Variabilidade permite estabelecer quais os
melhores par?metros para serem utilizados na Compara??o de Locutores. O ?ndice
permitiu ordenar as vogais e tamb?m embasou a proposta de que as medidas de espa?o
ac?stico e de Dist?ncias Euclidianas s?o poss?veis par?metros a serem utilizados na
Compara??o de Locutores. Diferentemente da Fon?tica Ac?stica e Articulat?ria, que
fundamenta a pesquisa, a Fonologia Articulat?ria comp?e o trabalho como aporte te?rico
acess?rio em rela??o a interpreta??o dos dados, por?m entende-se que seja teoria
valiosa para explica??o dos dados lingu?sticos que usualmente encontram-se na pratica
forense. As Dist?ncias Euclidianas entre os v?rtices do tri?ngulo voc?lico ? [a], [i] e [u] ?
serviram para propor uma an?lise de como a implementa??o computacional do modelo
da Din?mica de Tarefas, o TADA, trata vogais de forma cont?nua e consoantes de forma
categ?rica.
| Identifer | oai:union.ndltd.org:IBICT/oai:tede2.pucrs.br:tede/7798 |
| Date | 16 October 2017 |
| Creators | Pinto, M?rcio Oppliger |
| Contributors | Brescancini, Cl?udia Regina |
| Publisher | Pontif?cia Universidade Cat?lica do Rio Grande do Sul, Programa de P?s-Gradua??o em Letras, PUCRS, Brasil, Escola de Humanidades |
| Source Sets | IBICT Brazilian ETDs |
| Language | Portuguese |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, info:eu-repo/semantics/doctoralThesis |
| Format | application/pdf |
| Source | reponame:Biblioteca Digital de Teses e Dissertações da PUC_RS, instname:Pontifícia Universidade Católica do Rio Grande do Sul, instacron:PUC_RS |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | 8447345070736321569, 500, 500, 500, -6557531471218110192, -5409419262886498088 |
Page generated in 0.003 seconds