Dissertation
Dissertation > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Electro-acoustic technology and speech signal processing > Speech Signal Processing

Research on NetEQ Technology in WebRTC Voice Engine

Author WuJiangRui
Tutor ZhangGangShan; WangJing
School Xi'an University of Electronic Science and Technology
Course Electronics and Communication Engineering
Keywords VOIP NetEQ Jitter elimination Packet Lost Concealment Speech Quality
CLC TN912.3
Type Master's thesis
Year 2013
Downloads 37
Quotes 0
Download Dissertation

With the rapid development of the computer network, VoIP telephone service hasalready been widely used. This voice communication technology which based on IPnetwork provides a kind of convenient and cheap voice calls, and brings great changesto communication industry. But, the requirement of voice quality which appearssubsequently in VoIP, is urgent.WebRTC is one of the most advanced voice engine technology, and contains theNetEQ module, which provides complicated speech signal processing. The NetEQmodule provides a good solution to those problems in VoIP system, such as delay, jitterand packet loss, so widely used. But the NetEQ is a patent technology, and there is nopublic document to introduce or describe its detailed algorithm. Thus, a research andanalysis of NetEQ is necessary for better application of it in VoIP industry.The NetEQ algorithm integrated the adaptive jitter elimination algorithm and voicepacket loss concealment algorithm, which is the core contents of this paper. Firstly, thesystem structure and processing procedure of NetEQ have been analyzed briefly in thispaper. Then, the emphasis is putted on the control mechanism of NetEQ’s controlmodule(MCU) and the speech data processing method of processing module(DSP). Theanalysis results showed that the NetEQ indeed has an advantage in jitter elimination andpacket lost concealment.Finally, a speech quality evaluation module, which based on the speech quality ofNetEQ, is developed in this paper. The evaluation module is based on voice qualityprediction model (E-Model) and single-end non-intrusive speech quality measurement(P.563), and an UML sequence diagram to describe development processes and methodsof the speech quality evaluation module is also provide in this paper. The evaluationresults are mainly in agreement with the subjective feeling.

Related Dissertations
More Dissertations