Speech recognition using neural networks with forward-backward probability generated targets
Lists
Only you see this
Abstract
Neural network training targets for speech recognition are estimated using a novel method. Rather than use zero and one, continuous targets are generated using forward-backward probabilities. Each training pattern has more than one class active. Experiments showed that the new method effectively decreased the error rate by 15% in a continuous digits recognition task.
BibTeX
@inproceedings{yan1997speech,
author = {Yonghong Yan and Mark A. Fanty and Ronald A. Cole},
title = {{Speech recognition using neural networks with forward-backward probability generated targets}},
booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
year = {1997},
url = {https://openreview.net/forum?id=cnLanViRac},
}Loading related content...