Predictive model for determining the quality of a call

M. Voznak; J. Rozhon; P. Partila; J. Safarik; M. Mikulec; M. Mehic

doi:10.1117/12.2050661

22 May 2014 Predictive model for determining the quality of a call

M. Voznak, J. Rozhon, P. Partila, J. Safarik, M. Mikulec, M. Mehic

Proceedings Volume 9118, Independent Component Analyses, Compressive Sampling, Wavelets, Neural Net, Biosystems, and Nanoengineering XII; 91180Y (2014) https://doi.org/10.1117/12.2050661
Event: SPIE Sensing Technology + Applications, 2014, Baltimore, MD, United States

Abstract

In this paper the predictive model for speech quality estimation is described. This model allows its user to gain the information about the speech quality in VoIP networks without the need of performing the actual call and the consecutive time consuming sound file evaluation. This rapidly increases usability of the speech quality measurement especially in high load networks, where the actual processing of all calls is rendered difficult or even impossible. This model can reach its results that are highly conformant with the PESQ algorithm only based on the network state parameters that are easily obtainable by the commonly used software tools. Experiments were carried out to investigate whether different languages (English, Czech) have an effect on perceived voice quality for the same network conditions and the language factor was incorporated directly into the model.

Citation Download Citation

M. Voznak, J. Rozhon, P. Partila, J. Safarik, M. Mikulec, and M. Mehic "Predictive model for determining the quality of a call", Proc. SPIE 9118, Independent Component Analyses, Compressive Sampling, Wavelets, Neural Net, Biosystems, and Nanoengineering XII, 91180Y (22 May 2014); https://doi.org/10.1117/12.2050661

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available