Home // MOBILITY 2012, The Second International Conference on Mobile Services, Resources, and Users // View article

Speech Quality Assessment in Mobile Phones Using a Reduced- Complexity Algorithm

Authors:
Akram Aburas
Khalid Al-Mashouq
Musharraf Maqbool

Keywords: Speech Quality Measurement; Perceptual Evaluation of Speech Quality (PESQ); Signal Strength; Bit Error Rate (BER); Frame Erasure Rate (FER); Neural network.

Abstract:
In this paper, we present a reduced-complexity algorithm to assess the quality of speech as perceived by mobile user. This algorithm utilizes the channel parameters, as measured by the mobile handset, to estimate the speech quality. We used two estimation models; a linear model and a neural network-based model. We compared our estimation with the standard International Telecommunication Union, ITU, objective speech quality measure, perceptual evaluation of speech quality. We found that the linear model can achieve up to eighty four percent correlations with perceptual evaluation of speech quality. Moreover, the neural network-based model can achieve more than Ninety percent correlations with perceptual evaluation of speech quality.

Pages: 51 to 54

Copyright: Copyright (c) IARIA, 2012

Publication date: October 21, 2012

Published in: conference

ISSN: 2308-3468

ISBN: 978-1-61208-229-5

Location: Venice, Italy

Dates: from October 21, 2012 to October 26, 2012