Implementation of PCA & ICA for Voice ecognition and Separation of Speech

摘要：

Principle Component Analysis is great to evaluate the correlation among variable and reduce data dimensionally without loss of any data. The ability of analyzing the property of voice, reducing noises and extracting the valuable data of voice makes PCA an integral part of voice recognition. In digital signal processing signal estimation is required; signal may be superimposed by several interfering sources. To find one desired source signal Independent Component Analysis can be implemented. ICA recovers a set of independent signal from a set of measured signals by using statistical analysis of signal.

作者: A.Nitin Kandpal B.B.Madhusudan Rao

作者单位: Sost Department I2IT Pune, India Embedded Department KIT Pune, India

会议类型: 国际会议

会议名称: 2010 IEEE International Conference on Advanced Management Science(2010年IEEE高级管理科学国际会 IEEE ICAMS 2010)

会议地点: 成都

会议语种:英文

页码: 536-538

在线出版日期: 2010-07-09（万方平台首次上网日期，不代表论文的发表时间）