Análisis de datos de voz en personas autistas

Ruiz Cerdá, Álvar (2020). Análisis de datos de voz en personas autistas. Proyecto Fin de Carrera / Trabajo Fin de Grado, E.T.S. de Ingenieros Informáticos (UPM), Madrid, España.

Description

Title: Análisis de datos de voz en personas autistas
Author/s:
  • Ruiz Cerdá, Álvar
Contributor/s:
  • Rodellar Biarge, Victoria
Item Type: Final Project
Degree: Grado en Ingeniería Informática
Date: July 2020
Subjects:
Freetext Keywords: TEA; Parámetros de la voz; Temblor; Prosodia; Fuente glotal ASD; Voice parameters; Tremor; Prosody; Glotal font
Faculty: E.T.S. de Ingenieros Informáticos (UPM)
Department: Arquitectura y Tecnología de Sistemas Informáticos
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img]
Preview
PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (5MB) | Preview

Abstract

El trastorno del espectro autista (TEA) es una patología neurológica que se caracteriza por presentar graves déficits en el desarrollo de la persona. Estos trastornos afectan a la comunicación, a la interacción social, a la planificación, a la reciprocidad emocional, y a veces se manifiestan como conductas repetitivas o inusuales. Una de las mayores dificultades a las que se enfrentan los médicos es a un habitual retraso en el diagnóstico, debido fundamentalmente a la variabilidad individual que presenta esta condición. El presente trabajo es prospectivo y ha consistido en analizar registros de voz de personas autistas y con distintos grados de discapacidad. Se ha estudiado individualmente a los pacientes a través de dos muestras separadas por 6 meses. Los resultados obtenidos se compararán con los del resto de sujetos. Se han extraído los parámetros de la voz y la identificación de aquellas características que están alteradas respecto a la población normal. Los resultados obtenidos indican una alteración clara en algunos de los parámetros (biomecánicos y temblor), siendo comunes en todos los pacientes con TEA y discapacidad. Este trabajo es pionero en la búsqueda de parámetros y promete ser un interesante comienzo de una nueva línea de investigación.---ABSTRACT---The autism spectrum or autistic spectrum disorder (ASD), describes a neurological pathology that is characterised by the appearance of severe dysfunctions on the development of the person. These disorders affect communication, social interaction, the planning process and emotional reciprocity, and in certain circumstances, also appear as repetitive or restrictive behaviours. One of the biggest challenges doctors are faced up to is the fact that diagnosis is often made belatedly. This is essentially due to the fact that the disease manifests individual variability. The purpose of this research is to be future-oriented and it has consisted on analysing a display of voice registers belonging to people that suffer from autism on its different levels of affectation. Patients have been considered and studied individually through a sample-drawing procedure in a sixth-month period. The results obtained were compared to the rest of subjects that had been studied. Voice parameters have been extracted and additionally, those voice characteristics that are disrupted in comparison with the ones of normal population, have been identified. The analysis of biometric parameters regarding voice features, let us dig deeper into parameters closely related to ASD and, what is more, parameters that have in common with people that do not manifest this pathology. Furthermore, these parameters contain the speaker’s identity as well as a trace analysis of his or her mental emotional and neurological state. The achieved results show a clear alteration in some of the parameters, being found in all patients with ASD. This research leads the way in the pursuit of voice characteristic parameters in people that suffer from ASD and it is considered to be a great starting point for new lines of investigation in people with ASD.

More information

Item ID: 65206
DC Identifier: http://oa.upm.es/65206/
OAI Identifier: oai:oa.upm.es:65206
Deposited by: Biblioteca Facultad de Informatica
Deposited on: 03 Nov 2020 13:36
Last Modified: 03 Nov 2020 13:36
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM