Voice conversion using deep neural networks with speaker-independent pre-training.

Seyed Hamidreza Mohammadi Alexander Kain

Published in: SLT (2014)

Keyphrases

neural network
training process
speaker independent
speech recognition
pattern recognition
hidden markov models
speaker dependent
back propagation
speech recognizer
acoustic models
training set
training data
multilayer perceptron
discriminative training
digit recognition