Title :
Speech Independent Component Analysis
Author :
Wei, Hong ; Shi, Xinling ; Yang, Jian ; Pu, Yuanyuan
Author_Institution :
Sch. of Inf. Sci. & Technol., Yunnan Univ., Kunming, China
Abstract :
Speech is assumed that its production come from some different sources. Based on the independent component analysis, we separate its independent component from a speech and discuss the correlation and formants traits of these independent component signals. The synthesized speech is produced with the independent component features, and the power spectrum of these independent components is proved to be complimentary by the experimental results.
Keywords :
independent component analysis; speech synthesis; LPC power spectrum; speech independent component analysis; speech synthesis; Automation; Databases; Frequency; Hidden Markov models; Independent component analysis; Information science; Mechatronics; Signal processing algorithms; Speech analysis; Speech synthesis; Correlation; Formants; ICA; LPC power spectrum; Speech synthesis;
Conference_Titel :
Measuring Technology and Mechatronics Automation (ICMTMA), 2010 International Conference on
Conference_Location :
Changsha City
Print_ISBN :
978-1-4244-5001-5
Electronic_ISBN :
978-1-4244-5739-7
DOI :
10.1109/ICMTMA.2010.604
