After pointed the unreasonableness of the three basic assumptions contained in HMM, we introduce the theory and the advantage of Stochastic najectory Models (STMs) that possibly resolve these problems caused by HMM as...After pointed the unreasonableness of the three basic assumptions contained in HMM, we introduce the theory and the advantage of Stochastic najectory Models (STMs) that possibly resolve these problems caused by HMM assumptions. In STM, the acoustic observations of an acoustic unit are represented as clusters of trajectories in a parameter space.The trajectories are modelled by mixture of probability density functions of random sequence of states. After analyzing the characteristics of Chinese speech, the acoustic units for continuous Chinese speech recognition based on STM are discussed and phone-like units are suggested. The performance of continuous Chinese speech recognition based on STM is studied on VINICS system. The experimental results prove the efficiency of STM and the consistency of phone-like units.展开更多
This paper presents a new discriminative approach for training Gaussian mixture models(GMMs)of hidden Markov models(HMMs)based acoustic model in a large vocabulary continuous speech recognition(LVCSR)system.This appro...This paper presents a new discriminative approach for training Gaussian mixture models(GMMs)of hidden Markov models(HMMs)based acoustic model in a large vocabulary continuous speech recognition(LVCSR)system.This approach is featured by embedding a rival penalized competitive learning(RPCL)mechanism on the level of hidden Markov states.For every input,the correct identity state,called winner and obtained by the Viterbi force alignment,is enhanced to describe this input while its most competitive rival is penalized by de-learning,which makes GMMs-based states become more discriminative.Without the extensive computing burden required by typical discriminative learning methods for one-pass recognition of the training set,the new approach saves computing costs considerably.Experiments show that the proposed method has a good convergence with better performances than the classical maximum likelihood estimation(MLE)based method.Comparing with two conventional discriminative methods,the proposed method demonstrates improved generalization ability,especially when the test set is not well matched with the training set.展开更多
文摘After pointed the unreasonableness of the three basic assumptions contained in HMM, we introduce the theory and the advantage of Stochastic najectory Models (STMs) that possibly resolve these problems caused by HMM assumptions. In STM, the acoustic observations of an acoustic unit are represented as clusters of trajectories in a parameter space.The trajectories are modelled by mixture of probability density functions of random sequence of states. After analyzing the characteristics of Chinese speech, the acoustic units for continuous Chinese speech recognition based on STM are discussed and phone-like units are suggested. The performance of continuous Chinese speech recognition based on STM is studied on VINICS system. The experimental results prove the efficiency of STM and the consistency of phone-like units.
基金The work was supported in part by the National Natural Science Foundation of China(Grant No.90920302)the National Key Basic Research Program of China(No.2009CB825404)+2 种基金the HGJ Grant(No.2011ZX01042-001-001)a research program from Microsoft China,and by a GRF grant from the Research Grant Council of Hong Kong SAR(CUHK 4180/10E)Lei XU is also supported by Chang Jiang Scholars Program,Chinese Ministry of Education for Chang Jiang Chair Professorship in Peking University.
文摘This paper presents a new discriminative approach for training Gaussian mixture models(GMMs)of hidden Markov models(HMMs)based acoustic model in a large vocabulary continuous speech recognition(LVCSR)system.This approach is featured by embedding a rival penalized competitive learning(RPCL)mechanism on the level of hidden Markov states.For every input,the correct identity state,called winner and obtained by the Viterbi force alignment,is enhanced to describe this input while its most competitive rival is penalized by de-learning,which makes GMMs-based states become more discriminative.Without the extensive computing burden required by typical discriminative learning methods for one-pass recognition of the training set,the new approach saves computing costs considerably.Experiments show that the proposed method has a good convergence with better performances than the classical maximum likelihood estimation(MLE)based method.Comparing with two conventional discriminative methods,the proposed method demonstrates improved generalization ability,especially when the test set is not well matched with the training set.