| 基于压缩感知的语音信号编码算法 |
| |
| 引用本文： | 王茂林,黄文明,王菊娇.基于压缩感知的语音信号编码算法[J].桂林电子科技大学学报,2012(4):293-297. |
| |
| 作者姓名： | 王茂林 黄文明 王菊娇 |
| |
| 作者单位： | 桂林电子科技大学计算机科学与工程学院;广西交通职业技术学院信息工程系 |
| |
| 摘 要： | 针对语音信号在离散余弦变换基上的稀疏性,提出了一种基于压缩感知的语音压缩编码算法。算法在编码端采用随机高斯矩阵直接对语音波形进行观测,并采样均匀量化技术对随机观测进行量化。解码端利用未饱和的观测值通过Lasso算法实现语音信号的重构。仿真结果表明,该算法具有良好的重构性能。
|
| 关 键 词： | 压缩感知 语音编码 Lasso算法 |
A speech coding algorithm based on compressed sensing |
| |
| Wang Maolin,Huang Wenming,Wang Jujiao.A speech coding algorithm based on compressed sensing[J].Journal of Guilin Institute of Electronic Technology,2012(4):293-297. |
| |
| Authors: | Wang Maolin Huang Wenming Wang Jujiao |
| |
| Affiliation: | 1.School of Computer Science and Engineering,Guilin University of Electronic Technolgy,Ginlin 541004,China; 2.Department of Information Engineering,Guangxi Vocational and Technical College of Communications,Nanning 530023,China) |
| |
| Abstract: | A new speech coding algorithm based on compressed sensing is presented for the speech signal sparse representation with discrete cosine transform(DCT).This algorithm uses Gaussian random matrix for the speech waveform measurement,which is quantized independently by uniform quantizer.Saturated measurements are simply discarded in the decoder,then speech signals are reconstructed on those remained measurements by Lasso algorithm.Experimental results show that the performance of the new algorithm is good. |
| |
| Keywords: | compressed sensing speech coding Lasso algorithm |
| 本文献已被 CNKI 等数据库收录！ |
