一种基于混合MELP/CELP的4 kbit/s声码器  

A 4 kbit/s vocoder based on hybrid MELP/CELP

作　　者：李强[1] 舒勤军 谢虹恩 明艳[1] 

机构地区：[1]重庆邮电大学信号与信息处理重庆市重点实验室,重庆400065

出　　处：《重庆邮电大学学报（自然科学版）》2017年第2期143-148,共6页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：国家高技术研究发展计划(863计划)(2012AA01A508)~~

摘　　要：利用混合激励线性预测(mixed excitation linear prediction,MELP)算法和码激励线性预测(code excitation linear prediction,CELP)算法的优点,提出了一种混合MELP/CELP语音编码模型。编码端对强浊音帧采用MELP编码,对弱浊音帧和清音帧进行CELP编码。MELP编码器采用相位对齐技术提取强浊音帧的相位参数,解决了合成语音与原始语音在时间上不同步的问题。对实现的4 kbit/s混合MELP/CELP声码器进行客观MOS(mean opinion score)值和主观DRT(diagnostic rhythm test)清晰度测试,结果表明,该声码器的合成语音具有较高的可懂度和清晰度。Taking advantages of mixed excitation linear prediction（MELP） algorithm and code excitation linear prediction （CELP） algorithm, a hybrid MELP/CELP algorithm model is proposed. At the encoder, strongly voiced frames are coded by MELP while weakly voiced frames and unvoiced frames are coded by CELP. In MELP vocoder, the phase alignment technology is adopted to deal with the synchronization issue of original speech and synthesized speech. Objective mean opinion score（MOS） test and subjective diagnostic rhythm test（DRT） on 4 kbit/s hybrid MELP vocoder results indicate that the proposed algorithm offers good performance on high frequency speech signal and the synthesized speech, which achieves high intelligibility and clarity.

关 键 词：低码率声码器 混合激励线性预测 码激励线性预测 相位对齐 

分 类 号：TN912.3[电子电信—通信与信息系统]