الفهرس 
chapter 1Only 14 pages are availabe for public view
 |  | from | 161 |  |  |
| | | from | 161 | | |
Abstract
In the realm of wireless ommunications, voice has always been on of the most
dominant services for the very simple reason that speech is the most desirable
means of communication between people .Due to a limited frequency
spectrum, voice communication systems need to incorporate speech coding
techniques to save bandwidth. Conventional methods like Pulse Code
Modulation (PCM), Delta Modulation (DM),etc .. , are not suitable as their
bandwidth occupancy is comparatively high.
Speech coding techniques are classified into three main groups: Waveform
coders, parametric coders and Hybrid coders. Waveform coders offer
excellent voice quality with higher bit rates. Parametric coders assumed that
speech signal is generated from a model are estimated from the input speech
signal this class of coder has the lower bit rate but less voice quality. Hybrid coders combine the advantages of waveform coders and parametric coders.
The goal of this thesis is to implement evaluation criterion to assess the quality performance of advanced hybrid speech coding techniques in time domain,
spectral domain and perceptual domain by using Matlab simulation program.
This thesis evaluates three algorithms of advanced hybrid speech coding
technique that are commonly applied in today’s applications. The three
algorithms are (CELP), (G729-ANNEX A) and (G723.1) to assess the quality
performance for English female speaker, English male speaker and Arabic
female speaker by using Matlab simulation program.
Analytical evaluations are applied for three algorithms to measure speech
Quality in terms of bit rates they work on using SNR, SSNR, Absolute Error,
Mean Square Error, PESQ, LLR,WSS, Rating of speech distortion, rating of
background noise and the predicted rating of overall quality.