Title :
Robust voice conversion systems using MFDWC
Author :
Farhid, M. ; Tinati, M.A.
Author_Institution :
Univ. of Tabriz, Tabriz
Abstract :
Voice conversion is a method used to transform one speakerpsilas voice into another speakerpsilas voice. New modification approach for voice conversion is proposed in this paper. We take Mel-frequency Discrete Wavelet coefficients (MFDWC) as the basic feature. This feature copes well with small training sets of high dimension, which is a problem often encountered in voice conversion. The proposed voice conversion system assumes parallel training data from source and target speakers and uses the theory of wavelets in order to extract speaker feature information. The satisfactory performance of the voice conversion system can be confirmed through ABX listening test and MOS grade.
Keywords :
discrete wavelet transforms; feature extraction; speaker recognition; ABX listening test; MFDWC; MOS grade; mel-frequency discrete wavelet coefficients; parallel training data; speaker feature information; voice conversion systems; wavelets theory; Data mining; Discrete wavelet transforms; Feature extraction; Frequency; Linear predictive coding; Loudspeakers; Robustness; Speech analysis; Speech synthesis; Wavelet coefficients;
Conference_Titel :
Telecommunications, 2008. IST 2008. International Symposium on
Conference_Location :
Tehran
Print_ISBN :
978-1-4244-2750-5
Electronic_ISBN :
978-1-4244-2751-2
DOI :
10.1109/ISTEL.2008.4651405
