Abstract

Summary In recent years, there has been a trend towards training large vocabulary continuous speech recognition (LVCSR) systems on a large amount of found data. Found data is recorded from spontaneous speech without careful control of the recording acoustic conditions, for example, conversational telephone speech. Hence, it typically has greater variability in terms of speaker and acoustic conditions than specially collected data. Thus, in addition to the desired speech variability required to discriminate between words, it also includes various non-speech variabil-ities, for example, the change of speakers or acoustic environments. The standard approach to handle this type of data is to train hidden Markov models (HMMs) on the whole data set as if all data comes from a single acoustic condition. This is referred to as multi-style training, for exam-ple speaker-independent training. Effectively, the non-speech variabilities are ignored. Though good performance has been obtained with multi-style systems, these systems account for all variabilities. Improvement may be obtained if the two types of variabilities in the found data are modelled separately. Adaptive training has been proposed for this purpose. In contrast to multi-style training, a set of transforms is used to represent the non-speech variabilities. A canonical

Citations

7510	Maximum likelihood from incomplete data via the em algorithm - Dempster, Laird, et al. - 1976
2438	Introduction to statistical pattern recognition. 2 nd edition - Fukunaga - 1990
1769	Pattern Classification - Duda, Hart, et al. - 2001
1358	Fundamentals of Speech Recognition - Rabiner, Juang - 1993
1172	Statistical Decision Theory and Bayesian Analysis - BERGER - 1985
1077	Error bounds for convolutional codes and an asymptotically optimum decoding algorithm - Viterbi - 1967
1058	Pattern Recognition and Neural Networks - Ripley - 1996
814	Monte Carlo statistical methods - Robert, Castella - 2004
681	Comparison of Parametric Representations for Monosyllabic Word Recognition in Continuously Spoken Sentences - Davis, Mermelstein - 1980
636	Optimal Statistical Decisions - DeGroot - 1970
531	Maximum likelihood linear regression for speaker adaptation of continuous density hidden Markov models - Leggetter, Woodland - 1995
447	A model for reasoning about persistence and causation - Dean, Kanazawa - 1989
446	Maximum a posteriori estimation for multivariate Gaussian mixture observation of Markov chains - Gauvain, Lee - 1994
373	Maximum likelihood linear transformations for HMMbased speech recognition - Gales - 1998
336	The population frequencies of species and the estimation of population parameters - Good - 1953
252	Variational algorithms for approximate Bayesian inference - Beal - 2011
216	The zero-frequency problem: Estimating the probabilities of novel events in adaptive text compression - Witten, Bell - 1991
203	An inequality with applications to statistical estimation for probabilistic functions for a Markov process and to a model for ecology - Baum, Egon - 1967
188	Some statistical issues in the comparison of speech recognition algorithms - Gillick, Cox - 1989
187	Accurate approximations for posterior moments and marginal densities - Tierney, Kadane - 1986
174	Continuous Speech Recognition by Statistical Methods - Jelinek - 1976
171	Semi-tied covariance matrices for hidden Markov models - Gales - 1999
171	Tree-based state tying for high accuracy acoustic modeling - Young, Odell, et al. - 1994
168	Minimum Phone Error and I-smoothing for Improved Discriminative Training - Povey, Woodland - 2002
167	Effectiveness of linear prediction characteristics of the speech wave for automatic speaker identification and verification - Atal - 1974
150	Maximum mutual information estimation of Hidden Markov Model parameters for speech recognition - Bahl, Brown, et al. - 1986
135	A compact model for speaker-adaptive training - Anastasakos, McDonough, et al. - 1996
116	Minimum classification error rate methods for speech recognition - Juang, Chou, et al.
107	Discriminative Training for large Vocabulary Speech Recognition - Povey - 2004
96	Large scale discriminative training of hidden Markov models for speech recognition - Woodland, Povey - 2002
94	Mean and variance adaptation within the MLLR framework,” Computer Speech and Language - Gales, Woodland - 1996
91	Confidence measures for large vocabulary continuous speech recognition - Wessel, Schlüter, et al. - 2001
84	Rapid speaker adaptation in Eigenvoice space - Kuhn, Junqua, et al. - 2000
83	The N-best algorithm: An efficient and exact procedure for finding the N most likely sentence hypotheses - SCHWARTZ, CHOW - 1990
82	Speaker adaptation using constrained estimation of gaussian mixtures - Digalakis, Rtischev, et al. - 1995
78	Speaker normalization using efficient frequency warping procedures - Lee, Rose - 1996
73	1’hc dragon system an overview - Baker - 1975
72	Flexible speaker adaptation using maximum likelihood linear regression - Leggetter, Woodland - 1995
71	Investigation of Silicon-Auditory Models and Generalization of Linear Discriminant Analysis for Improved Speech Recognition - Kumar - 1997
65	Large Scale Discriminative Training for Speech Recognition - Woodland, Povey - 2000
64	Structural maximum a posteriori linear regression for fast HMM adaptation, Computer Speech and Language - Siohan, Myrvoll, et al. - 1994
64	An empirical Bayes approach to statistics - Robbins - 1955
60	Maximum likelihood estimation for multivariate mixture observations of Markov chains - Juang, Levinson, et al. - 1986
59	The development of the 1994 HTK large vocabulary speech recognition system - Woodland, Leggetter, et al. - 1995
57	The generation and use of regression class trees for MLLR adaptation - Gales - 1996
53	Minimum Bayes-risk automatic speech recognition - Goel, Byrne - 2000
52	Cluster adaptive training of hidden markov models - Gales - 2000
52	Large vocabulary continuous speech recognition using HTK - Woodland, Odell, et al. - 1994
49	RASTA-PLP speech analysis technique - Hermansky, Morgan, et al. - 1992
43	Using word probabilities as confidence measures - Wessel, Macherey, et al. - 1998