Mapping features using Deep Neural Networks (DNNs) with application to Voice Conversion (VC). The implementations are on top of Theano Python library Includes:

- Deep Neural Networks with Dropout

- Stacked Denoising Autoencoders

Requirements:

1- numpy 2.7

2- theano 0.6

Please refer to this paper to cite this work:

S.H. Mohammadi, A. Kain, Semi-supervised Training of a Voice Conversion Mapping Function using Joint-Autoencoder, Interspeech (To Appear), 2015.

S.H. Mohammadi, A. Kain, Voice Conversion Using Deep Neural Networks With Speaker-Independent Pre-Training, 2014 IEEE Spoken Language Technology Workshop (SLT), 2014.