Title :
Improving the Performance of GMM Based Voice Conversion Method
Author :
Song, Peng ; Zhao, Li
Author_Institution :
Sch. of Inf. Sci. & Eng., Southeast Univ., Nanjing
Abstract :
This paper describes a modified GMM based voice conversion method,in order to avoid the over-smoothed phenomena of traditional GMM conversion method, we adopt GMM method considering posterior threshold of each frame. We also considered the dynamic spectral features between frames to overcome the speech discontinuity problem,experiments show that our presented GMM method has shown better performance than traditional one.
Keywords :
Gaussian processes; regression analysis; speech processing; GMM based voice conversion method; dynamic spectral features; linear multivariate regression method; speech discontinuity problem; Computational intelligence; Computer industry; Conferences; Information science; Land mobile radio; Mean square error methods; Multivariate regression; Speech synthesis; Statistical analysis; Vector quantization; GMM; dynamic features; posterior threshold;
Conference_Titel :
Computational Intelligence and Industrial Application, 2008. PACIIA '08. Pacific-Asia Workshop on
Conference_Location :
Wuhan
Print_ISBN :
978-0-7695-3490-9
DOI :
10.1109/PACIIA.2008.34
