Towards end-to-end F0 voice conversion
based on Dual-GAN with convolutionnal wavelet kernels
IRCAM, CNRS, Sorbonne Université
STMS Lab, Paris, France
Clément LE MOINE, Nicolas OBIN and Axel ROEBEL
Dual-GAN ugraded with PreNet based convolutionnal wavelet kernels
The proposed system is a end-to-end framework for the F0 transformation in the context of expressive voice conversion. A single neural network is proposed, in which a first module is used to learn F0 representation over different temporal scales and a second adversarial module is used to learn the transformation from one attitude to another. The first module is composed of a convolution layer with wavelet kernels so that the various temporal scales of F0 variations can be efficiently encoded. The single decomposition/transformation network allows to learn in a end-to-end manner the F0 decomposition that are optimal with respect to the transformation, directly from the raw F0 signal.
Paper
Clément LE MOINE, Nicolas OBIN, Axel ROEBEL, "Towards end-to-end F0 voice conversion based on Dual-GAN with convolutionnal wavelet kernels" ResearchGate
Voice attitudes conversion examples
1. Friendy - Distant
2. Friendly - Seductive
3. Distant - Seductive
4. Distant - Dominant
5. Dominant - Seductive
6. Friendly - Dominant