Font Size: a A A

G.729 Speech Co-decoding Algorithm Investigation And Implementation Based On DSP

Posted on:2009-06-07Degree:MasterType:Thesis
Country:ChinaCandidate:P LiuFull Text:PDF
GTID:2178360272987330Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
The fundamental of speech coding is the fact that how to get better quality of reconstructed speech and ensure the less time delay and proper operational complexity at the given coding rate. G.729 standard that was produced by ITU-T is a conjugate-structure algebraic code excited linear prediction low-rate speech coding project, which is commonly abbreviated CS-ACELP. G.729 standard has been widely used on VOIP gateway and IP phone because of its high nicer synthesized speech quality and appropriate complexity and less time delay. G.729 has already became one of the best important modern communication speech coding projects.This thesis first introduces the fundamental of low-rate speech coding, and specifies G.729 standard in detail, analyzes the important technologies in theory (include linear prediction, vector quantization, codebook search, post-filter and so on). Moreover, speech co-decoder based on G.729 is lucubrated in the thesis. After being familiar with C language standard source code of G.729 Annex A, the author advanced efficient implementation of G.729 co-decoder which based on the platform of TMS320C5416 EMV board that was produced by ChangChengJiaXin Co. Under the condition of guaranteeing favorable speech quality, it is necessary to optimize the algorithm reduce the code quantity after algorithm of ITU-T recommendation G.729 be analysed. After the algorithm is optimized, the amended code is transplanted to the platform of TMS320C5416 EMV, then structure G.729 speech Co-decoder. Then, the performance of its code is evaluated by the profiler which is a tool of analyzing code, and the result indicate that G.729 arithmetic can be implement in real time on a single C5416. The test result shows the singnal from this system has good natural characteristic, definition and intelligibility, and its reconstructed sound is similar to the original one. The reconstructed singnal wave is similar to the original one, and it retains whole information of the original signal approximately.
Keywords/Search Tags:G.729, CS-ACELP, low-rate speech coding, TMS320C5416
PDF Full Text Request
Related items