Speech recognition using neural networks with forward-backward probability generated targets

Yonghong Yan, Mark A. Fanty, Ronald A. Cole

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.