Speech restoration based on deep learning autoencoder with layer-wised pretraining.

Xugang Lu Shigeki Matsuda Chiori Hori Hideki Kashioka

Published in: INTERSPEECH (2012)

Keyphrases

deep learning
restricted boltzmann machine
unsupervised learning
machine learning
unsupervised feature learning
image processing
multiscale
mental models
information retrieval
image segmentation
object recognition
natural images
weakly supervised
deep architectures