Dissertation > Excellent graduate degree dissertation topics show
Research on Mixed Excitation Linear Prediction Speech Coding Algorithm
Author: FanJing
Tutor: HeYingMin
School: Harbin Engineering University
Course: Signal and Information Processing
Keywords: Speech coding MELP Mixed excitation Pitch period LSF
CLC: TN912.3
Type: Master's thesis
Year: 2008
Downloads: 106
Quote: 1
Read: Download Dissertation
Abstract
|
With the rapid development of communications technology , and real-time transmission of Internet voice technology , voice transfer rate and storage capacity of high demands, one of the main ways to solve these problems is a speech coding . Therefore , the study of speech coding , especially in low bit rate speech coding , has a very important practical significance . In existing speech coding , the mixed excitation linear predictive speech coding (MELP) is a better method , which combines the binary excitation the code excitation and multi-band excitation advantage , the short-time speech segment is divided into a number of sub-band , respectively , in each sub-band unvoiced discriminating ; in the synthesis side , the mixing sequence to a sequence of periodic pulses and random noise excitation speech synthesis filter , in the lower bit rate to get a better regeneration voice . The 2.4kbps mixed excitation linear predictive speech coding has been established as the new U.S. Federal speech coding standards . The dissertation research MELP speech codec algorithm principle , its codec to do a more in-depth research , we can also make certain improvements in the pitch period and the linear prediction coefficients quantify this . In the standard MELP algorithm , for the voice signal segment contain irregular cycles , the calculated mutual correlation value is small, and mistakenly believe that it is unvoiced , and therefore introduces noise . Here used a method of changing the algorithm of the pitch period , the more accurate calculation of the pitch period . In addition, in the process of LSF quantization , the codebook storage amount and complexity of computation are great. For this problem , we propose three vector quantization method , which can MELP bit rate drops about 2.1kbps , there is still a good quality of synthetic speech . Finally, the MATLAB programming environment the MELP algorithm and its improved the MELP algorithm simulation , the simulation results show that after the decoded voice signal and improved voice signal output waveform of the original speech signal waveform is very similar , but in energy voiced synthesized speech waveform relatively large amplitude and delay , in order to verify the feasibility of the proposed algorithm .
|
Related Dissertations
- The Studt of ILBC Speech Coding’s Key Technology and Its DSP Design and Optimization,TN912.3
- Research on Very Low Bit Rate Speech Coding,TN912.3
- Research and Optimization of Job Scheduling in LSF System,TP311.52
- Real-Time Performance Reseach on Embedded Meteorologic Acquisition System and Network,TP274.2
- The Implementation and Improvement of MELP Speech Coding Algorithm,TN912.3
- Hybrid Excitation Brushless AC synchronous motor excitation converter of,TM341
- Implementation of CS-ACELP Algorithm on TMS320C54X,TN912.3
- Implementation of G.728 Speech Coder Based on TMS320C64X,TN912.3
- Research on Low Bit Rates Waveform Interpolation Speech Coding Based on Singular Value Decomposition,TN912.3
- Low Delay Vector Exciting Speech Coding Algorithm Based on BP Neural Network,TN912.3
- Mixed excitation MVDR speech coding,TN912.3
- Hardware Design and Implementation of the simulation system based on DSP and Internet technology VHF,U676.2
- Low- rate speech coding algorithm based on linear predictive coding research,TN912.3
- The voice companding algorithm and hardware implementation,TN912.3
- Research and DSP Implementation on High Quality Speech Coding Algorithm at 4KB/S,TN912.3
- Research of Low Bit-Rate Sinusoidal Transform Coding,TN911.22
- Implementation of Speech Compress Coding Based on DSP,TN912.3
- Research of G.729 Speech Coder Implementation Based on DSP,TN912.3
- G.729D Speech Codec Implementation on TMS320C64X,TN912.3
- The Research and Development of the Embedded Speech System Based on MC68HC12,TN912.3
CLC: > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Electro-acoustic technology and speech signal processing > Speech Signal Processing
© 2012 www.DissertationTopic.Net Mobile
|