摘要
压缩感知,近年来在信号处理领域取得非常多的成果。在基于压缩感知进行语音压缩重构时,测量矩阵在重构原始信号的过程中发挥着重要的作用.为了提高重构语音质量,从观测矩阵着手,基于紧框架算法构造测量矩阵,并与传统的高斯随机观测矩阵进行重构语音质量的比较。实验结果表明,紧框架矩阵相对于传统常用的高斯随机矩阵,在语音重构过程中取得更好的效果。
Compressed sensing is a research hotspot making so much progress in the field of signal processing. We know that the observation matrix plays an important role in the reconstruction of original speech signal. In order to improve the quality of reconstructed speech, starts from the observation matrix, constructs observation matrix based on tight frame algorithm and compares with the traditional gaussian random observation matrix on reconstructing of original speech. The experiments＇ results show that the tight frame matrix, compared with the traditional gaussian random matrix achieves better effect in the process of speech re-construction.
出处
《现代计算机》
2016年第5期31-35,共5页
Modern Computer
关键词
语音
压缩感知
紧框架
观测矩阵
高斯随机矩阵
Speech
Compressed Sensing
Tight Frame
Observation Matrix
Gaussian Random Matrix
