Mandarin phoneme independent component analysis and synthesis

Author

Hong Wei ; Jian Yang

Author_Institution

Sch. of Inf. Sci. & Technol., Yunnan Univ., Kunming, China

Volume

03

fYear

2013

fDate

16-18 Dec. 2013

Firstpage

1314

Lastpage

1318

Abstract

Speech is assumed that its production come from some different vocal sources. Regarding the Chinese mandarin phonemes, based on the independent component analysis, we separate its temporal independent components from a specific phoneme and discuss the formants, fundamental frequency F0, linear predictive coding (LPC) spectrum and Wigner-Ville (WV) spectrum of every independent component. Based on the LPC and Wigner-Ville spectrum windowed in the formants position, the different timbre phoneme is synthesized. Our analysis is proved to be effective by the experimental results.

Keywords

independent component analysis; linear predictive coding; natural language processing; speech synthesis; LPC; Mandarin phoneme independent component analysis; Wigner-Ville spectrum; fundamental frequency; linear predictive coding spectrum; speech synthesis; temporal independent components; timbre phoneme; Cavity resonators; Hidden Markov models; Independent component analysis; Signal processing algorithms; Speech; Speech synthesis; Timbre; F0; LPC spectrum; Wigner-Ville spectrum; formants; independent components; speech synthesis;

fLanguage

English

Publisher

ieee

Conference_Titel

Image and Signal Processing (CISP), 2013 6th International Congress on

Conference_Location

Hangzhou

Print_ISBN

978-1-4799-2763-0

Type

conf

DOI

10.1109/CISP.2013.6743876

Filename

6743876