Abstract
In this paper, two main shortcomings of the traditional Gaussian mixture model (GMM) based spectral transformation method are addressed. One is the over-smoothing caused by the statistic averaging of the model, the other is the discontinuities because of the frame-wise conversion. Aiming at compensating for these two problems, a novel spectral transformation method based on clustering and regression is proposed to solve the over-smoothing. Meanwhile, a novel averaging strategy is adopted to reduce the discontinuities. In order to further improve the perceptual speech quality, being different from the traditional methods, a novel F0 transformation method combining the F0 prediction with the Gaussian normalization is presented. Objective and subjective experiments are carried out to demonstrate the efficiency of the proposed method.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Abe, M., Nakanura, S., Shikano, K., Kuwabara, H.: Voice Conversion Through Vector Quantization. In: 13th International Conference on Acoustics, Speech, and Signal Processing, pp. 655–658. IEEE Press, New York (1988)
Kain, A., Macon, M.W.: Spectral Voice Conversion for Text-to-Speech Synthesis. In: 23th International Conference on Acoustics, Speech, and Signal Processing, pp. 285–288. IEEE Press, Seattle (1998)
Desai, S., Black, A.W., Yegnanarayana, B., Prahallad, K.: Spectral Mapping Using Artificial Neural Networks for Voice Conversion. IEEE Transactions on Audio, Speech, and Language Processing 18(5), 954–964 (2010)
Chen, Y., Chu, M., Chang, E., Liu, J., Liu, R.: Voice Conversion with Smoothed GMM and MAP Adaptation. In: 8th European Conference on Speech Communication and Technology, pp. 2413–2416. ISCA, Geneva (2003)
Toda, T., Black, A.W., Tokuda, K.: Spectral Conversion Based on Maximum Likelihood Estimation Considering Global Variance of Converted Parameter. In: 30th International Conference on Acoustics, Speech, and Signal Processing, pp. 9–12. IEEE Press, Philadelphia (2005)
Inanoglu, Z.: Transforming Pitch in a Voice Conversion Framework. Master thesis. St Edmund’s College, University of Cambridge, Cambridge (2003)
Pérez-Cruz, F., Camps-Valls, G., Soria-Olivas, E., Pérez-Ruixo, J.J., Figueiras-Vidal, A.R., Artes-Rodriguez, A.: Multi-dimensional Function Approximation and Regression Estimation. In: 12th International Conference on Artificial Neural Networks, pp. 757–762. Springer, Madrid (2002)
Song, P., Bao, Y.Q., Zhao, L., Zou, C.R.: Voice Conversion Using Support Vector Regression. Electronics Letters 47(18), 1045–1046 (2011)
Reynolds, D.A., Quatieri, T.F., Dunn, R.B.: Speaker Verification Using Adapted Gaussian Mixture Models. Digital Signal Processing 10(1), 19–41 (2000)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Song, P., Bao, Y., Zhao, L. (2012). Voice Conversion Using Improved Spectral and F0 Transformation Methods. In: Liu, CL., Zhang, C., Wang, L. (eds) Pattern Recognition. CCPR 2012. Communications in Computer and Information Science, vol 321. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-33506-8_72
Download citation
DOI: https://doi.org/10.1007/978-3-642-33506-8_72
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-33505-1
Online ISBN: 978-3-642-33506-8
eBook Packages: Computer ScienceComputer Science (R0)