Calvo-Ariza NR, Arias-Vergara T, Orozco-Arroyave JR (2023)
Publication Type: Conference contribution
Publication year: 2023
Publisher: Springer Science and Business Media Deutschland GmbH
Book Volume: 1928 CCIS
Pages Range: 127-138
Conference Proceedings Title: Communications in Computer and Information Science
Event location: Cartagena, COL
ISBN: 9783031467387
DOI: 10.1007/978-3-031-46739-4_12
The early detection of pathologies is a significant challenge in the healthcare sector and often requires the involvement of multiple specialists in different areas. The analysis of bio-signals, such as speech, gait, and video, has shown promising results in detecting pathologies. In this study, we propose using phase plots extracted from electroglottographs (EGG) and acoustic signals to detect voice disorders automatically. The method consists in using the visual information provided by the plots and processing the information with convolutional neural networks for quantitative analysis of voice quality. We test our approach with EGG and acoustic signals from the Saarbruecken Voice Database (SVD), which contains recordings of 687 controls and 1186 patients diagnosed with 71 different voice pathologies. In addition to the phase plots, state-of-the-art features were computed for quantitative assessment. An 85 % accuracy was obtained when we combined the proposed phase plots with filter-bank features.
APA:
Calvo-Ariza, N.R., Arias-Vergara, T., & Orozco-Arroyave, J.R. (2023). Automatic Assessment of Voice Disorders Using Phase Plots. In Juan Carlos Figueroa-García, Elvis Eduardo Gaona García, Jose Luis Villa Ramirez, German Hernández (Eds.), Communications in Computer and Information Science (pp. 127-138). Cartagena, COL: Springer Science and Business Media Deutschland GmbH.
MLA:
Calvo-Ariza, N. R., T. Arias-Vergara, and J. R. Orozco-Arroyave. "Automatic Assessment of Voice Disorders Using Phase Plots." Proceedings of the 10th Workshop on Engineering Applications, WEA 2023, Cartagena, COL Ed. Juan Carlos Figueroa-García, Elvis Eduardo Gaona García, Jose Luis Villa Ramirez, German Hernández, Springer Science and Business Media Deutschland GmbH, 2023. 127-138.
BibTeX: Download