The Research And Design Of Variable Rate Speech Coding Based On DSP

To meet the needs of the third mobile communication system for higher speech quality and greater system capacity, the CDMA systems employ variable rate speech coding technologies, which select the encode rate dynamically according to the signal energy and background noise.This method insures the speech quality as well as reduces the average encode rate, so the system capacity is increase. The enhanced variable rate codec's the maximal encode rate is 8kbps, its voice quality is closed to QCELP-13k and has beter ability of anti-disturbance. Enhanced Variable Rate Codec (EVRC) is one of the most applied variable rate speech coding technologies.This paper introduces the research and design of EVRC speech coding Based on DSP. Firstly, this paper introduces the study actuality and development trend of variable rate speech coding.Secondly, this paper expounds the relative theory and technologies of speech coding, at the same time, the key to technologies used in variable rate speech coding are studied. They include Voice Activity Detection , Rate Decision Algorithm, Erasure Concealment Unit and Comfort Noise Aspects.And then,EVRC algorithm is concentrate on studing in this paper,The basic principle of main function modules are analyzed and explained systemically .Moreover some pivotal technologies are analyzed in detail.Thirdly,this paper introduces the structure and the work principle for the tms320vc5409 used in this system.On the base of theory analysis,the project of hardware compose is putted forword about whole system.Finally,the collectivity design idea of sofeware is expatiated.Besides,this paper discusses Rate Decision Algorithm and Fixed Codebook Search in all rate.Meantime, some modifications are suggested.The function of whole algorithm is realized using C language. In the programming process, a series of optimization work is performed to implement the algorithm better.the system is simulated and tested on software level,.The subjective and objective performance of main function modules is given.Through simulated and tested result, the system can availability implement encoding and decoding to the real-time voice.Quality of voice reconstructed is well understood. Systemic performance is satisfied for factual requirements.