基于压缩感知的语音增强识别系统.pdfVIP

Vol. 33 No.2 第 33 卷第 2 期 计算机应用研究 2016 年2 月 Application Research of Computers Feb. 2016 基于压缩感知的语音增强识别系统* 茅正冲，龚熙 (江南大学轻工过程先进控制教育部重点实验室，江苏无锡214122) 摘 要:针对传统压缩感知(CS) 方法抗噪性能差的问题，提出了一种新的基于压缩感知的语音增强识别系统。 该系统在用正交匹配追踪(OMP) 算法重构语音信号时设定相关皮阅位和语音恢复闽佳，并对迭代算法进行改 进，不仅有效恢复了纯净语音信号，实现了语音增强，并且减少了重构的计算量;再将重构恢复的信号通过Gam­ matone 滤泼器组提取特征参数GFCC ，并在高斯混合模型中匹配。 仿真实验表明，将这种方法应用于卢纹识别系 统，系统的识别率a鲁棒性都有明显提高。 关键词:压缩感知;正交匹配追踪;抗嗓算法;识别系统 中图分类号: TN912.3 文黯标志码:A 文章编号: 1∞1-3695(2016)02-0450-04 doi: 10. 3969/j. issn. 1∞1-3695.2016.02.031 Speech enhancement recognition system based on CS Mao Zhengchong. Gong Xi (Key μboratory 01Adt跑时ed Prooω ContTollor l号hllnd旧町• 01Ministry 01Educat阳. JiangTUln Uni时时町• WuxiJiangsu 214122. ChiTUl) Abstract: In view of the p∞r anti-noise perfonnance of traditional compressive sensing ,this paper proposed a novel speech enhancement recognition system bωed on compressive sensing. The proposed system set similarity threshold and speech signal recoveηthreshold in orthogonal matching persuit ( OMP) algorithm. and improved the iterative algorithm ,which not only re- stored the enhanced speech signal ,but also reduced the amount of calculation. Then it extracted feature p缸ameters CFCC of the enhanced speech si伊al by Cammatone filter bank ,and matched the best result in Gaussian mixture model. Simulation experi- ments show that this method obviously improves the recognition rate and robustness in speaker recognition system. Key words: compressive sensing(CS); orthogonal matching pursuit(OMP); anti-noise algorithm; recognition system