Banca de QUALIFICAÇÃO: WINNIE DE LIMA TORRES

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
DISCENTE : WINNIE DE LIMA TORRES
DATA : 20/06/2017
HORA: 10:00
LOCAL: Laboratório de Automação, Controle e Instrumentação (LACI)
TÍTULO:

Classification of Vocal Disturbances Using System Identification

PALAVRAS-CHAVES:

Laryngeal Pathologies, Systems Identification, Classification of Vocal Disturbances

PÁGINAS: 52
GRANDE ÁREA: Engenharias
ÁREA: Engenharia Elétrica
SUBÁREA: Eletrônica Industrial, Sistemas e Controles Eletrônicos
ESPECIALIDADE: Controle de Processos Eletrônicos, Retroalimentação
RESUMO:

Several fields of science propose to study disorders in the vocal tract from analyzes on patterns of voice vibration. In general, the importance of these researches is in the identification, at a preliminary stage, of diseases of greater or lesser severity, to be healed with vocal therapy or that require more attention, even generating the need for surgical procedures for its control. Although there are indications in the literature that the digital signal processing allows non-invasive diagnosis of laryngeal pathologies such as vocal diseases that cause: edema, nodule and paralysis, there is no definition of the most indicated method and the characteristics or parameters most adequate to detect the presence of vocal deviations. In this way, this work seeks to test the techniques of identification of systems for the modeling of voice signals, in order to develop an algorithm that allows the classification of normal and pathological voices, thus contributing to the studies carried out in the area, in particular in the field of identification of systems applied to biomedical signals. To perform the research, we used constant data in the Disordered Voice Database, developed by the Massachusetts Eye and Ear Infirmary (MEEI), due to the credibility of the source and its wide use in research in the acoustic area of voice. We used 166 signals, with duration of 1 to 3 seconds and frequency of 25 kHz contained in this database, with signs of healthy voices and pathological voices affected by edema, nodule and paralysis in the vocal folds. Initially, we performed basic statistical analyzes and establishment of routines that made it possible to verify if voice signals can only be classified with a measure of dissimilarity or if they require application of other methodologies to identify and classify laryngeal pathologies. It was found that only the basic statistical analysis is still insufficient to achieve the proposed objective, the identification and classification of vocal disorders. Thus, we proceeded to a more specific phase, applying methodologies to identify systems to detect the structure and estimation of parameters using an AR model. This process provided similar results for the methodologies (CGS, MGS and FROLS) for identifying systems addressed in the work.

MEMBROS DA BANCA:
Presidente - 347565 - ALDAYR DANTAS DE ARAUJO
Interno - 2579664 - ALLAN DE MEDEIROS MARTINS
Externo à Instituição - ADEMAR GONÇALVES DA COSTA JÚNIOR - IFPB