Publication detail

Voice Pathology Detection Using Deep Learning: a Preliminary Study

HARÁR, P. ALONSO-HERNANDEZ, J. MEKYSKA, J. GALÁŽ, Z. BURGET, R. SMÉKAL, Z.

Original Title

Voice Pathology Detection Using Deep Learning: a Preliminary Study

Type

conference paper

Language

English

Original Abstract

This paper describes a preliminary investigation of Voice Pathology Detection using Deep Neural Networks (DNN). We used voice recordings of sustained vowel /a/ produced at normal pitch from German corpus Saarbruecken Voice Database (SVD). This corpus contains voice recordings and electroglottograph signals of more than 2 000 speakters. The idea behind this experiment is the use of convolutional layers in combination with recurrent Long-Short-Term-Memory (LSTM) layers on raw audio signal. Each recording was split into 64 ms Hamming windowed segments with 30 ms overlap. Our trained model achieved 71.36% accuracy with 65.04% sensitivity and 77.67% specificity on 206 validation files and 68.08% accuracy with 66.75% sensitivity and 77.89% specificity on 874 testing files. This is a promising result in favor of this approach because it is comparable to similar previously published experiment that used different methodology. Further investigation is needed to achieve the state-of-the-art results.

Keywords

voice pathology detection; deep learning; convolutional layers; long-short-term-memory layers; audio signal

Authors

HARÁR, P.; ALONSO-HERNANDEZ, J.; MEKYSKA, J.; GALÁŽ, Z.; BURGET, R.; SMÉKAL, Z.

Released

24. 7. 2017

Location

Funchal, Portugal

ISBN

978-1-5386-0850-0

Book

2017 International Work Conference on Bio-inspired Intelligence (IWOBI)

Pages from

45

Pages to

48

Pages count

4

BibTex

@inproceedings{BUT138230,
  author="HARÁR, P. and ALONSO-HERNANDEZ, J. and MEKYSKA, J. and GALÁŽ, Z. and BURGET, R. and SMÉKAL, Z.",
  title="Voice Pathology Detection Using Deep Learning: a Preliminary Study",
  booktitle="2017 International Work Conference on Bio-inspired Intelligence (IWOBI)",
  year="2017",
  pages="45--48",
  address="Funchal, Portugal",
  doi="10.1109/IWOBI.2017.7985525",
  isbn="978-1-5386-0850-0"
}