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

Research and Implementation of Adaptive Low-Rate Speech Coding

Author ZhouXiaoHai
Tutor WangMin
School Xi'an University of Architecture and Technology
Course Control Theory and Control Engineering
Keywords speech coding linear prediction coding AMR endpoint detection CELP DSP
CLC TN912.3
Type Master's thesis
Year 2009
Downloads 27
Quotes 0
Download Dissertation

Speeching communication is the main service that the mobile communication network can provide now.original digital speech takes up more bandwidth,but the capability of the mobile communication network is limited,each channel that allocated to user is also limited,so it is possible to compress original digital speech before to be transmitted.Because the adjacent sampling points are strongly correlative,it is viable to compress digital speech.Speech coding includes parameter coding and wave figure coding.There are many advantages for using parameter coding.including low coding rate,excellent quality,excellent anti-noise performance and etc.so it is widely applied in most speech coding standard now.with the increase of mobile telephone user,it is necessary to split the difference between communication network capability and user number,so modifiable coding rate become the main coding mode that widely used in mobile communication network.Adaptive Multi-Rate,a speech coding standard that constituted by ITU is used in 3G mobile communication network,it is based on CELP and can improve high quality compositive speech.AMR can adjust coding mode and coding rate by the change of channel and resource.So it can improve efficiency of channel and optimize the capability of mobile communication network,indicating the development direction of speech coding,the key technology of AMR includes voice activate detection,RAD and CANIn the chapter one,the paper introduces the basal knowledge of speech coding and actuality in the world,in the chapter two,the paper introduces the basal knowledge of speech coder,linear prediction based on parameter coding and a improved endpoint detection algorithm,since the chapter three,the paper detailedly introduces the components of AMR,the algorithms of encoder and decoder,the key technologies of AMR,an improved method of backdrop noise prediction,in chapter four to chapter three,the paper introduces the hardware design of AMR based on TMS320VC5402 DSP,and get the result of hardware experiment and software experiment.

Related Dissertations
More Dissertations