基于最小熵的音频信号高频重建  

Minimum Entropy Based High Frequency Reconstruction of Audio Signal

作　　者：白晓亮[1] 胡剑凌[1] 徐盛[1] 

机构地区：[1]上海交通大学图像通信与信息处理研究所,上海200030

出　　处：《电声技术》2006年第2期37-40,共4页Audio Engineering

摘　　要：音频信号的有损压缩编码为实现高的压缩比,往往会将高频部分删除,这样势必导致音质下降。为此,很多研究工作围绕恢复高频展开,其技术统称为高频重建。提出了一种基于最小熵的方法进行高频重建,试验结果表明重建声音主观听觉良好。In audio coding, the high frequency components of signals will be removed for higher compression ratio, hence lead to great degradation of quality. Therefore, a new technique called high frequency reconstruction （HFR） comes out to improved the quality of reconstructed sound. This paper introduces a novd approach of HFR, which is based on minimum entropy. Experimental results show that the subjective testing of reconstructed sound is good.

关 键 词：数字音频 高频重建 最小熵 谱估计 

分 类 号：TN912[电子电信—通信与信息系统]