Deep-learning in Identification of Vocal Pathologies

Felipe Teixeira, João Teixeira

Abstract

The work consists in a classification problem of four classes of vocal pathologies using one Deep Neural Network. Three groups of features extracted from speech of subjects with Dysphonia, Vocal Fold Paralysis, Laryngitis Chronica and controls were experimented. The best group of features are related with the source: relative jitter, relative shimmer, and HNR. A Deep Neural Network architecture with two levels were experimented. The first level consists in 7 estimators and second level a decision maker. In second level of the Deep Neural Network an accuracy of 39,5% is reached for a diagnosis among the 4 classes under analysis.

Download


Paper Citation


in Harvard Style

Teixeira F. and Teixeira J. (2020). Deep-learning in Identification of Vocal Pathologies.In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 4: BIOSIGNALS, ISBN 978-989-758-398-8, pages 288-295. DOI: 10.5220/0009148802880295


in Bibtex Style

@conference{biosignals20,
author={Felipe Teixeira and João Teixeira},
title={Deep-learning in Identification of Vocal Pathologies},
booktitle={Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 4: BIOSIGNALS,},
year={2020},
pages={288-295},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0009148802880295},
isbn={978-989-758-398-8},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 4: BIOSIGNALS,
TI - Deep-learning in Identification of Vocal Pathologies
SN - 978-989-758-398-8
AU - Teixeira F.
AU - Teixeira J.
PY - 2020
SP - 288
EP - 295
DO - 10.5220/0009148802880295