Voice-based medical diagnostic device development

OData support
Supervisor:
Dr. Vicsi Klára
Department of Telecommunications and Media Informatics

At the beginning of the semester the purpose was to implement a real time application that is able to calculate and visualize the main acoustic parameters of human voice. These parameters are the followings: pitch, jitter, shimmer, harmonics-to-noise ratio.

In the course of the variant medical examination it has been found that the values of these parameters are considerably distinct in the case of healthy and pathological voices. This could be observed especially by determining the values of jitter, shimmer and harmonics-to-noise ratio.

The purpose of my application is to give a sort of base for the user to locate how the calculated values of his/her voice relate to the typical values of healthy voice.

The work started by learning about the forenamed parameters, followed by working on the algorithms that are used to determine these factors.

To check the operability of the implemented application I compared my results with the effects of another speech diagnostic software which does not operate in real time. Generally the result was satisfying; the values given by the two programs were similar.

As the last phase of the work with the implemented application I analyzed the voice of eight male and eight female volunteers.

Downloads

Please sign in to download the files of this thesis.