ByoVoz Automatic Voice Condition Analysis System for the 2018 FEMH Challenge

Arias-Londoño, Julián David, Gómez García, Jorge Andrés ORCID: https://orcid.org/0000-0002-6060-387X, Moro Velázquez, Laureano and Godino Llorente, Juan Ignacio ORCID: https://orcid.org/0000-0001-7348-3291 (2018). ByoVoz Automatic Voice Condition Analysis System for the 2018 FEMH Challenge. In: "Proceedings IEEE International Conference on Big Data 2018", 10/12/2018 -13/12/2018, Seatle, WA, USA. ISBN 978-1-5386-5035-6. pp. 5228-5232. https://doi.org/10.1109/BigData.2018.8622498.

Description

Title: ByoVoz Automatic Voice Condition Analysis System for the 2018 FEMH Challenge
Author/s:
Item Type: Presentation at Congress or Conference (Article)
Event Title: Proceedings IEEE International Conference on Big Data 2018
Event Dates: 10/12/2018 -13/12/2018
Event Location: Seatle, WA, USA
Title of Book: 2018 IEEE International Conference on Big Data (Big Data)
Date: December 2018
ISBN: 978-1-5386-5035-6
Subjects:
Freetext Keywords: Voice pathology detection; Voice pathology identification; Gradient boosting; Gaussian mixture models; Random forest
Faculty: E.T.S.I. y Sistemas de Telecomunicación (UPM)
Department: Teoría de la Señal y Comunicaciones
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[thumbnail of INVE_MEM_2018_300984.pdf]
Preview
PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (302kB) | Preview

Abstract

This paper presents the methods and results used by the ByoVoz team for the design of an automatic voice condition analysis system, which was submitted to the 2018 Far East Memorial Hospital voice data challenge. The proposed methodology is based on a cascading scheme that firstly discriminates between pathological and normophonic voices, and then identifies the type of disorder. By using diverse feature selection techniques, a subset of complexity, spectral/cepstral and perturbation characteristics were identified for the proposed tasks. Then, several generative classification methodologies based on Gaussian Mixture Models and Gradient Boosting were employed to provide decisions about the input voices in the binary classification, and using onevs-one classification systems based on Random Forests for the categorization according to the type of disorder. By using a 4-folds cross-validation approach on the training partition a sensitivity=0.93 and specificity=0.74 were obtained. Similarly, an unweighted average recall of 0.63 and an accuracy of 66% was obtained for the identification task. Using the scoring metric proposed in the challenge the final resulting score considering both detection and identification is of 0.77.

Funding Projects

Type
Code
Acronym
Leader
Title
Government of Spain
DPI2017-83405-R
Unspecified
Ministerio de Economía, Industria y Competitividad
Biomarcadores para el diagnóstico y la evaluación de la enfermedad de Parkinson basados en estudios multimodales de voz y oculografía

More information

Item ID: 55117
DC Identifier: https://oa.upm.es/55117/
OAI Identifier: oai:oa.upm.es:55117
DOI: 10.1109/BigData.2018.8622498
Deposited by: Memoria Investigacion
Deposited on: 24 Feb 2020 14:25
Last Modified: 24 Feb 2020 14:25
  • 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