You are in:Home/Publications/Performance Evaluation of EVS Coder against the change in the Spoken Language

Assist. Christina Gamal :: Publications:

Performance Evaluation of EVS Coder against the change in the Spoken Language
Authors: Christina Gamal, Michael N.Mikhael, Hala A. Mansour
Year: 2017
Keywords: LTE, 3GPP, EVS, AMR, G.729, PESQ, MOSLQO, Speech quality
Journal: Not Available
Volume: Not Available
Issue: Not Available
Pages: Not Available
Publisher: Not Available
Local/International: Local
Paper Link: Not Available
Full paper Not Available
Supplementary materials Not Available

The quality of speech is essential in communication systems. Enhanced Voice Services (EVS) is one of the speech Coders that standardized by the 3GPP for LTE. EVS supports audio bandwidth, offers high quality for mixed content (i.e. speech and music). Many coders that depend on the concept of ACELP may be affected by the change in the spoken language or accent. EVS Coder working with both ACELP and MDCT coding techniques. This paper focuses on EVS performance evaluation against the change in the spoken language. Extensive testing is done on EVS, AMR-NB and G.729 Coders with 3 groups of speech databases (Arabic, American English, and British English). The speech quality is measured using PESQ and MOSLQO evaluation methods. It has been proven that for Arabic: EVS coder is better than both G.729 and AMR-NB respectively in terms of average speech quality and performance stability. It is also shown that the EVS gives the best quality and performance stability with American English.

Google ScholarAcdemia.eduResearch GateLinkedinFacebookTwitterGoogle PlusYoutubeWordpressInstagramMendeleyZoteroEvernoteORCIDScopus