Simulation and Implementation of Enhanced LPC Vocoder

Abstract

Speech coding has been and still is a major issue in the area of digital speech processing. Reducing the transmission bandwidth and achieving higher speech quality are primary concerns in developing new speech coding algorithms. The standard PCM produce voice signal with data rate of 64Kbps which reduces the efficiency of bandwidth utilization, increases the cost and complicates to use in security applications. The goal of this thesis is to develop a system for encoding good quality speech at low bit rate.To implement this, used very efficient speech analysis technique called Linear Predictive Coding (LPC). The basic idea behind linear predictive analysis is that the speech sample can be approximated as a linear combination of past samples. In this project, the simulation of the LPC vocoder was done using matlab. The output voice quality was tested by using perceptual evaluation signal quality (PESQ) and it was 60%.Then LPC was implemented in DSK C6713 chip, and the Mean Opinion Score (MOS) was used to evaluate the quality, the average rate of the MOS was 3 which means the quality is fair.