一种改进的2.4kb/s混合激励线性预测声码器方案  

An improved 2.4kb/s mixed excitation linear prediction vocoder

作　　者：马欣[1] 刘常澍[1] 李文元[1] 张毓忠[1] 

机构地区：[1]天津大学电子信息工程学院,天津300072

出　　处：《电路与系统学报》2007年第3期117-120,共4页Journal of Circuits and Systems

摘　　要：本文针对标准的2.4kb/s MELP声码器的不足之处提出了两项改进措施,一是提出了一种新的参数“能量—微分过零率比”,用来对语音的过渡段和弱能量浊音段的清浊音判决进行调整;二是对线谱对的多级矢量量化(MSVQ)提出了一种多径搜索算法。实验和主观听觉测试表明,在同样2.4kb/s的码率下,改进MELP声码器的合成语音在可懂度和自然度方面都有一定的提高。In this paper, two improvements on 2.4kb/s Mixed-Excitation linear Prediction （MELP） vocoder are presented. The one is a new parameter named Energy to Differential Zero-crossing Rate which is used in adaptation of V/UV decision of transitional segments and low energy level speech segments. The second is a multi-path searching method for multi-stage vector quantization （MSVQ） of line spectral frequency. Subjective tests show that the intelligiblity and naturallity of improved MELP vocoder are preferable to that of the original one.

关 键 词：清浊音判决 MELP 声码器 多级矢量量化(MSVQ) 

分 类 号：TN912.3[电子电信—通信与信息系统]