Deep-learning in identification of vocal pathologies

Teixeira, Felipe; Teixeira, João Paulo

http://hdl.handle.net/10198/26895

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
BIOSIGNALS_2020_42 (1)_SCITePRESS.pdf		521.41 KB	Adobe PDF	Download

Send Feedback

Authors

Teixeira, Felipe

Teixeira, João Paulo

Abstract(s)

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.

Keywords

Vocal acoustic analysis Leave-one-out Deep neural network Architecture of deep-NN Dysphonia Vocal fold paralysis Laryngitis chronica

URI

http://hdl.handle.net/10198/26895

Citation

Teixeira, Felipe; Teixeira, João Paulo (2020). Deep-learning in identification of vocal pathologies. In 13th International Joint Conference on Biomedical Engineering Systems and Technologies. Malta. ISSN 2184-4305. 4, p. 288-295.