This tutorial provides an overview of the basic theory of hidden Markov models (HMMs) as originated by L.E. Baum and T. Petrie (1966) and gives practical details on methods of implementation of the theory along with a description of selected applications of the theory to distinct problems in speech recognition. Results from a number of original sources are combined to provide a single source of acquiring the background required to pursue further this area of research. The author first reviews the theory of discrete Markov chains and shows how the concept of hidden states, where the observation is a probabilistic function of the state, can be used effectively. The theory is illustrated with two simple examples, namely coin-tossing, and the classic balls-in-urns system. Three fundamental problems of HMMs are noted and several practical techniques for solving these problems are given. The various types of HMMs that have been studied, including ergodic as well as left-right models, are described. >

/pdf/a-tutorial-on-hidden-markov-models-and-selected-applications-1vlzbxbk68.pdf

A tutorial on hidden Markov models and selected applications in speech recognition

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

We address the problem of predicting a word from previous words in a sample of text. In particular, we discuss n-gram models based on classes of words. We also discuss several statistical algorithms for assigning words to classes based on the frequency of their co-occurrence with other words. We find that we are able to extract classes that have the flavor of either syntactically based groupings or semantically based groupings, depending on the nature of the underlying statistics.

Class-based n -gram models of natural language

The authors present a time-delay neural network (TDNN) approach to phoneme recognition which is characterized by two important properties: (1) using a three-layer arrangement of simple computing units, a hierarchy can be constructed that allows for the formation of arbitrary nonlinear decision surfaces, which the TDNN learns automatically using error backpropagation; and (2) the time-delay arrangement enables the network to discover acoustic-phonetic features and the temporal relationships between them independently of position in time and therefore not blurred by temporal shifts in the input. As a recognition task, the speaker-dependent recognition of the phonemes B, D, and G in varying phonetic contexts was chosen. For comparison, several discrete hidden Markov models (HMM) were trained to perform the same task. Performance evaluation over 1946 testing tokens from three speakers showed that the TDNN achieves a recognition rate of 98.5% correct while the rate obtained by the best of the HMMs was only 93.7%. >

Phoneme recognition using time-delay neural networks

/pdf/phoneme-recognition-using-time-delay-neural-networks-3o10ydjad6.pdf

A description is presented of the authors' current research on automatic speech recognition of continuously read sentences from a naturally-occurring corpus: office correspondence. The recognition system combines features from their current isolated-word recognition system and from their previously developed continuous-speech recognition system. It consists of an acoustic processor, an acoustic channel model, a language model, and a linguistic decoder. Some new features in the recognizer relative to the isolated-word speech recognition system include the use of a fast match to prune rapidly to a manageable number the candidates considered by the detailed match, multiple pronunciations of all function words, and modeling of interphone coarticulatory behavior. The authors recorded training and test data from a set of ten male talkers. The perplexity of the test sentences was found to be 93; none of sentences was part of the data used to generate the language model. Preliminary (speaker-dependent) recognition results on these talkers yielded an average word error rate of 11.0%. >

/pdf/large-vocabulary-natural-language-continuous-speech-34gyw1bm42.pdf

Large vocabulary natural language continuous speech recognition

This paper studies distributed estimation of unstable dynamic random fields observed by a sparsely connected network of sensors. The field dynamics are globally detectable, but not necessarily locally detectable. We propose a consensus+innovations distributed estimator, termed Distributed Information Kalman Filter. We prove under what conditions this estimator is asymptotically unbiased with bounded mean-squared error, smaller than for other alternative distributed estimators. Monte Carlo simulations confirm our theoretical error asymptotic results.

Distributed Kalman Filtering With Dynamic Observations Consensus

In this paper, we address the distributed filtering and prediction of time-varying random fields represented by linear time-invariant (LTI) dynamical systems. The field is observed by a sparsely connected network of agents/sensors collaborating among themselves. We develop a Kalman filter type consensus + innovations distributed linear estimator of the dynamic field termed as Consensus+Innovations Kalman Filter. We analyze the convergence properties of this distributed estimator. We prove that the mean-squared error of the estimator asymptotically converges if the degree of instability of the field dynamics is within a prespecified threshold defined as tracking capacity of the estimator. The tracking capacity is a function of the local observation models and the agent communication network. We design the optimal consensus and innovation gain matrices yielding distributed estimates with minimized mean-squared error. Through numerical evaluations, we show that the distributed estimator with optimal gains converges faster and with approximately 3dB better mean-squared error performance than previous distributed estimators.

Consensus+Innovations Distributed Kalman Filter With Optimized Gains

The Speech Recognition Group at IBM Research in Yorktown Heights has developed a real-time, isolated-utterance speech recognizer for natural language based on the IBM Personal Computer AT and IBM Signal Processors. The system has recently been enhanced by expanding the vocabulary from 5,000 words to 20,000 words and by the addition of a speech workstation to support usability studies on document creation by voice. The system supports spelling and interactive personalization to augment the vocabularies. This paper describes the implementation, user interface, and comparative performance of the recognizer.

Experiments with the Tangora 20,000 word speech recognizer

There are several reasons why conventional speech recognition systems modeled on adult data fail to perform satisfactorily on children's speech input. For instance, children's vocal characteristics differ significantly from those of adults. In addition, their choices of vocabulary and sentence construction modalities usually do not conform to adult patterns. We describe comparative studies demonstrating the performance gain realized by adopting to children's acoustic and language model data to construct a children's speech recognition system.

Subhro Das

Papers

Large vocabulary natural language continuous speech recognition

Distributed Kalman Filtering With Dynamic Observations Consensus

Consensus+Innovations Distributed Kalman Filter With Optimized Gains

Experiments with the Tangora 20,000 word speech recognizer

Improvements in children's speech recognition performance