Diagnosing Voice Disorder with Machine Learning
Authors:
- 1 December 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5263-5266
Abstract
The goal of this study is to build an effective model to identify the types of voice disorder, include Normal, Neoplas, Phonotrauma and Vocal palsy, from FEMH dataset. To deal with this classification problem, we use several Machine learning methods, such as Support Vector Machine, Random Forest, K-Nearest Neighbor, and Gradient Boosting. Meanwhile, we also try the Ensemble Learning algorithm, which combines these learners and selects the most popular prediction.Keywords
This publication has 10 references indexed in Scilit:
- A Deep Learning Method for Pathological Voice Detection Using Convolutional Deep Belief Networks2018
- Modeling the Pathophysiology of Phonotraumatic Vocal Hyperfunction With a Triangular Glottal Model of the Vocal FoldsJournal of Speech, Language, and Hearing Research, 2017
- On combining information from modulation spectra and mel-frequency cepstral coefficients for automatic detection of pathological voicesLogopedics Phoniatrics Vocology, 2010
- Ensemble-based classifiersArtificial Intelligence Review, 2009
- Pathological voice discrimination using cepstral analysis, vector quantization and Hidden Markov Models2008
- Dimensionality Reduction of a Pathological Voice Quality Assessment System Based on Gaussian Mixture Models and Short-Term Cepstral ParametersIEEE Transactions on Biomedical Engineering, 2006
- Discrimination of Pathological Voices Using a Time-Frequency ApproachIEEE Transactions on Biomedical Engineering, 2005
- Support-vector networksMachine Learning, 1995
- An Introduction to Kernel and Nearest-Neighbor Nonparametric RegressionThe American Statistician, 1992
- The Generation of Confidence: Evaluating Research Findings by Random Subsample ReplicationSociological Methodology, 1972