New refinement schemes for voice conversion

Author

Lin, Cheng-Yuan ; Jang, J. S Roger

Author_Institution

Dept. of Comput. Sci., Nat. Tsing Hua Univ., Hsinchu, Taiwan

Volume

2

fYear

2003

fDate

6-9 July 2003

Abstract

New refinement schemes for voice conversion are proposed in this paper. We take mel-frequency cepstral coefficients (MFCC) as the basic feature and adopt cepstral mean subtraction to compensate the channel effects. We propose S/U/V (silence/unvoiced/voiced) decision rule such that two sets of codebooks are used to capture the difference between unvoiced and voiced segments of the source speaker. Moreover, we apply three schemes to refine the synthesized voice, including pitch refinement with PSOLA, energy equalization, and frame concatenation based on synchronized pitch marks. The satisfactory performance of the voice conversion system can be demonstrated through ABX listening test and MOS grade.

Keywords

cepstral analysis; speech processing; speech synthesis; cepstral mean subtraction; energy equalization; mel-frequency cepstral coefficients; synchronized pitch marks; unvoiced segments; voice conversion; voiced segments; Cepstral analysis; Computer science; Frequency conversion; Linear predictive coding; Loudspeakers; Mel frequency cepstral coefficient; Signal synthesis; Speech recognition; Speech synthesis; System testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Multimedia and Expo, 2003. ICME '03. Proceedings. 2003 International Conference on

Print_ISBN

0-7803-7965-9

Type

conf

DOI

10.1109/ICME.2003.1221719

Filename

1221719