A Discriminative Training Method Applied to a Hybrid ANN/HMM Phoneme Recognizer

摘要：

The challenge of this paper is to extend the concept of discriminative training to a hybrid ANN/HMM phoneme recognition system.The main goal is to improve phoneme accuracy in the aligned output string,instead of in the Multi Layer Perceptron (MLP) output,as usually done.The method uses the difference between the reference and the best acoustic likelihood of the observation sequences to update the MLP weights.Results are presented for the TIMIT phoneme recognition task and show that this method leads to significant improvements,comparing to the baseline.

作者: Carla Lopes Fernando Perdig(a)o

作者单位: Instituto de Telecomunica(c)(o)es;Department of Electrical and Computer Engineering,University of Co Instituto de Telecomunica(c)(o)es;Department of Electrical and Computer Engineering,University of Co

会议类型: 国际会议

会议名称: 9th International Conference on Signal Processing(第九届国际信号处理学术会议)(ICSP08)

会议地点: 北京

会议语种:英文

在线出版日期: 2008-10-26（万方平台首次上网日期，不代表论文的发表时间）