【作者】 李拟珺； 王仕奎； 孙明； 吴镇扬；

【Author】 LI Ni-jun~1 WANG Shi-kui~(1,2) SUN Ming~3 WU Zhen-yang~1 (1.School of Information Science and Engineering,Southeast University,Nanjing,Jiangsu,210096,China; 2.School of Physics and Electronic Information,Anhui Normal University,Wuhu,Anhui,241000,China; 3.Department of Electronic and Information Engineering,Foshan University,Foshan,Guangdong,528000,China)

【机构】 东南大学信息科学与工程学院； 安徽师范大学物理与电子信息学院； 佛山科学技术学院电子信息工程系；

【摘要】 由于ITU-T的G.722.2编码标准帧间激励的依赖性,导致其对丢失浊音端点十分敏感,会造成严重的错误传播。为改善其性能,设计了与G.722.2结合的宽带独立编码算法,它是由iLBC(internet Low Bit-rate Codec)改造的。提出了改进的算法,先用离散Teager能量算子对输入语音进行浊音端点检测,再对浊音端点附近进行独立编码以加强保护。实验结果表明,改进算法的平均码率在多数情况下略有提高,但消除了错误传播,在丢失较多浊音端点时的隐藏效果明显好于原算法。更多还原

【Abstract】 Due to the dependency between the excitation of adjacent frames of ITU-T G.722.2,the algorithm is subject to losing voiced onsets and will lead to severe error propagation.To improve its performance,the wideband independent codec algorithm which can be combined with G.722.2 is designed,and it is adapted from iLBC(internet Low Bit-rate Codec).An advanced algorithm is presented,which first uses discrete Teager energy operator to detect all voiced onsets of the input speech,and then encodes around the voiced onsets independently in order to protect them better.The experiment result indicates that the advanced algorithm has an average code rate slightly higher than the original algorithm in most cases,but it wipes out error propagation and has better concealment effects when losing more voiced onsets.更多还原