Staff Members - Benha University
Performance Evaluation of Speech CODECs against the Change in the Spoken Language and Accent
30 th NATIONAL RADIO SCIENCE CONFERENCE (NRSC 2013) • 2013
Publication Information
Authors
Michael N. Micheal, ,Nagy W. Messiha, , Hala A. Mansour
Keywords
performance, coders, accents, Arabic, Cairo accent, G.723.1, G.711, PESQ
Journal
30 th NATIONAL RADIO SCIENCE CONFERENCE (NRSC 2013)
Publisher
Not Available
Volume
30
Issue
URSI
Pages
Not Available
publication.type
International
Paper Link
Not Available
Supplementary Materials
Not Available
Abstract
In all modern communications, speech coding plays a very vital role. It is a basic element in the efficient use of
BW and QoS issues. Many standard bodies generate a series of speech coders with different rate, quality and
delay combinations. Most of these CODECs have been built for 7 languages not including the Arabic or its
accents. In this paper an extensive testing is done on the popular ITU G.723.1 and compared with the waveform
G711, µ_law. The test is done for (English, Arabic and Cairo accent) from different speakers’ gender. The
assessment were developed using the PESQ algorithm and it has been proved that the speech quality will be
slightly degraded when using other language or accent rather than English language. The main point is not only
the slight degradation but also the less stability of the coder performance when working with languages other than
English.
BW and QoS issues. Many standard bodies generate a series of speech coders with different rate, quality and
delay combinations. Most of these CODECs have been built for 7 languages not including the Arabic or its
accents. In this paper an extensive testing is done on the popular ITU G.723.1 and compared with the waveform
G711, µ_law. The test is done for (English, Arabic and Cairo accent) from different speakers’ gender. The
assessment were developed using the PESQ algorithm and it has been proved that the speech quality will be
slightly degraded when using other language or accent rather than English language. The main point is not only
the slight degradation but also the less stability of the coder performance when working with languages other than
English.