There are a lot of difficulties that have to be overcome in the speaker-independent (S.I.) phrase recognition system . And the feasibility of accurate ,real-time and robust system pose of the greatest challenges in the system. In this thesis ,the speaker-independent phase recognition system is based on Hidden Markov Model (HMM). HMM has been proved to be of great value in many applications, notably in speech recognition. HMM is a stochastic approach which characterizes many of the variability in speech signal. It applys the state-of-the-art approach to Automatic Speech Recognition .

A design of speech recognition system for Chinese names has been established in this thesis. By identifying surname first, that is an unique feature of the Chinese names, the classification accuracy and computational time of the system can be greatly improved. This research is primarily based on hidden Markov model (HMM), a technique that is widely used in speech recognition. HMM is a doubly stochastic process describing the ways of pronumciation by recording the state transitions according to the time-varing properties of the speech signal. The results of the HMM are compared with those of the segmental probability model (SPM) to figure out better option in recognizing base-syllables. Under the conditions of equal segments, SPM not only suits Mandarin base-syllable structure, but also achieves the goal of simplifying system since it does not need to find the best transformation of the utterance. A speaker-independent 3000 Chinese names recognition system has been implemented based on the Mandarin microphone database recorded in the laboratory environment.

We build a two-pass Mandarin Automatic Speech Recognition (ASR) decoder on mobile device (PDA). The first-pass recognizing base syllable is implemented by discrete Hidden Markov Model (HMM) with time-synchronous, tree-lexicon Viterbi search. The second-pass dealing with language model, pronunciation lexicon and N-best syllable hypotheses from first-pass is implemented by Weighted Finite State Transducer (WFST). The best word sequence is obtained by shortest path algorithms over the composition result. This system limits the application in travel domain and it decouples the application of acoustic model and the application of language model into independent recognition passes. We report the real-time recognition performance performed on ASUS P565 with a 800MHz processor, 128MB RAM running Microsoft Window Mobile 6 operating system. The 26-hour TCC-300 speech data is used to train 151 acoustic model. The 3-minute speech data recorded by reading the travel-domain transcriptions is used as the testing set for evaluating the performances (syllable, character accuracies) and real-time factors on PC and on PDA. The trained bi-gram model with 3500-word from BTEC corpus is used in second-pass. In the first-pass, the best syllable accuracy is 38.8% given 30-best syllable hypotheses using continuous HMM and 26-dimension feature. Under the above syllable hypotheses and acoustic model, we obtain 27.6% character accuracy on PC after the second-pass.

ITC/USA 2013 Conference Proceedings / The Forty-Ninth Annual International Telemetering Conference and Technical Exhibition / October 21-24, 2013 / Bally's Hotel & Convention Center, Las Vegas, NV / The transition of telemetry from link-based to network-based architectures opens these systems to new security risks. Tools such as intrusion detection systems and vulnerability scanners will be required for emerging telemetry networks. Intrusion detection systems protect networks against attacks that occur once the network boundary has been breached. An intrusion detection model was developed in the Wireless Networking and Security lab at Morgan State University. The model depends on network traffic being filtered into traffic streams. The streams are then reduced to vectors. The current state of the network can be determined using Viterbi analysis of the stream vectors. Viterbi uses the output of the Hidden Markov Model to find the current state of the network. The state information describes the probability of the network being in predefined normal or attack states based on training data. This output can be sent to a network administrator depending on threshold levels. In this project, a penetration-testing tool called Metasploit was used to launch attacks against systems in an isolated test bed. The network traffic generated during an attack was analyzed for use in the MSU intrusion detection model.

In telecommunication, the series of mobile network exceptions is a processwhich exhibits surges and bursts. The bursty part is usually caused by systemmalfunction. Additionally, the mobile network exceptions are often timedependent. A model that successfully captures these aspects will make troubleshootingmuch easier for system engineers. The Hidden Markov Model(HMM) is a good candidate as it provides a mechanism to capture both thetime dependency and the random occurrence of bursts. This thesis focuses onan application of the HMM to mobile network exceptions, with a case study ofEricsson’s Dropped Call data. For estimation purposes, two methods of maximumlikelihood estimation for HMM, namely, EM algorithm and stochasticEM algorithm, are used.

Philosophiae Doctor - PhD / In this thesis, we present a novel hybrid architecture by combining the most popular sequence recognition models such as Recurrent Neural Networks (RNNs) and Hidden Markov Models (HMMs). Though sequence recognition problems could be potentially modelled through well trained HMMs, they could not provide a reasonable solution to the complicated recognition problems. In contrast, the ability of RNNs to recognize the complex sequence recognition problems is known to be exceptionally good. It should be noted that in the past, methods for applying HMMs into RNNs have been developed by other researchers. However, to the best of our knowledge, no algorithm for processing HMMs through learning has been given. Taking advantage of the structural similarities of the architectural dynamics of the RNNs and HMMs, in this work we analyze the combination of these two systems into the hybrid architecture. To this end, the main objective of this study is to improve the sequence recognition/classi_cation performance by applying a hybrid neural/symbolic approach. In particular, trained HMMs are used as the initial symbolic domain theory and directly encoded into appropriate RNN architecture, meaning that the prior knowledge is processed through the training of RNNs. Proposed algorithm is then implemented on sample test beds and other real time biological applications.

The collection of large speech databases is not a trivial task (if done properly). It is not always possible to collect, segment and annotate large databases for every task or language. It is also often the case that there are imbalances in the databases, as a result of little data being available for a specific subset of individuals. An example of one such imbalance is the fact that there are often more male speakers than female speakers (or vice-versa). If there are, for example, far fewer female speakers than male speakers, then the recognizers will tend to work poorly for female speakers (as compared to performance for male speakers). This thesis focuses on using Bayesian and discriminative training algorithms to improve continuous speech recognition systems in scenarios where there is a limited amount of training data available. The research reported in this thesis can be divided into three categories: • Overspecialization is characterized by good recognition performance for the data used during training, but poor recognition performance for independent testing data. This is a problem when too little data is available for training purposes. Methods of reducing overspecialization in the minimum classification error algo¬rithm are therefore investigated. • Development of new Bayesian and discriminative adaptation/training techniques that can be used in situations where there is a small amount of data available. One example here is the situation where an imbalance in terms of numbers of male and female speakers exists and these techniques can be used to improve recognition performance for female speakers, while not decreasing recognition performance for the male speakers. • Bayesian learning, where Bayesian training is used to improve recognition perfor¬mance in situations where one can only use the limited training data available. These methods are extremely computationally expensive, but are justified by the improved recognition rates for certain tasks. This is, to the author's knowledge, the first time that Bayesian learning using Markov chain Monte Carlo methods have been used in hidden Markov model speech recognition. The algorithms proposed and reviewed are tested using three different datasets (TIMIT, TIDIGITS and SUNSpeech), with the tasks being connected digit recognition and con¬tinuous speech recognition. Results indicate that the proposed algorithms improve recognition performance significantly for situations where little training data is avail¬able. / Thesis (PhD (Electronic Engineering))--University of Pretoria, 2006. / Electrical, Electronic and Computer Engineering / unrestricted

Most of the chronic diseases have a well-known natural staging system through which the disease progression is interpreted. It is well established that the transition rates from one stage of disease to other stage can be modeled by multi state Markov models. But, it is also well known that the screening systems used to diagnose disease states may subject to error some times. In this study, a simulation study is conducted to illustrate the importance of addressing for misclassification in multi-state Markov models by evaluating and comparing the estimates for the disease progression Markov model with misclassification opposed to disease progression Markov model. Results of simulation study support that models not accounting for possible misclassification leads to bias. In order to illustrate method of accounting for misclassification is illustrated using dementia data which was staged as no cognitive impairment, mild cognitive impairment and dementia and diagnosis of dementia stage is prone to error sometimes. Subjects entered the study irrespective of their state of disease and were followed for one year and their disease state at follow up visit was recorded. This data is used to illustrate that application of multi state Markov model which is an example of Hidden Markov model in accounting for misclassification which is based on an assumption that the observed (misclassified) states conditionally depend on the underlying true disease states which follow the Markov process. The misclassification probabilities for all the allowed disease transitions were also estimated. The impact of misclassification on the effect of covariates is estimated by comparing the hazard ratios estimated by fitting data with progression multi state model and by fitting data with multi state model with misclassification which revealed that if misclassification has not been addressed the results are biased. Results suggest that the gene apoe ε4 is significantly associated with disease progression from mild cognitive impairment to dementia but, this effect was masked when general multi state Markov model was used. While there is no significant relation is found for other transitions.

In this thesis the Hidden Markov Model (HMM) is used in the process of map matching to investigate the accuracy for road segment map matching. A few HMM algorithms using a Global Navigation Satellite System (GNSS) receiver, odometer and gyroscope sensors are presented. The HMM algorithms are evaluated on four accuracy metrics. Two of these metrics have been seen in previous literature and captures road map match accuracy. The other have not been seen before and captures road segment accuracy. In the evaluation process a dataset is created by simulation to achieve positional ground truth for each sensor measurement. The accuracy distribution for different parts of the map matched trajectory is also evaluated. The result shows that HMM algorithms presented in previous literature, falls short to capture the accuracy for road segment map matching. The results further shows that by using less noisy sensors, as odometer and gyroscope, the accuracy for road segment map matching can be increased.

Dans cette thèse, nous nous intéressons à l'estimation de paramètres dans les chaînes de Markov cachées. Nous considérons tout d'abord le problème de l'estimation en ligne (sans sauvegarde des observations) au sens du maximum de vraisemblance. Nous proposons une nouvelle méthode basée sur l'algorithme Expectation Maximization appelée Block Online Expectation Maximization (BOEM). Cet algorithme est défini pour des chaînes de Markov cachées à espace d'état et espace d'observations généraux. Dans le cas d'espaces d'états généraux, l'algorithme BOEM requiert l'introduction de méthodes de Monte Carlo séquentielles pour approcher des espérances sous des lois de lissage. La convergence de l'algorithme nécessite alors un contrôle de la norme Lp de l'erreur d'approximation Monte Carlo explicite en le nombre d'observations et de particules. Une seconde partie de cette thèse se consacre à l'obtention de tels contrôles pour plusieurs méthodes de Monte Carlo séquentielles. Nous étudions enfin des applications de l'algorithme BOEM à des problèmes de cartographie et de localisation simultanées. La dernière partie de cette thèse est relative à l'estimation non paramétrique dans les chaînes de Markov cachées. Le problème considéré est abordé dans un cadre précis. Nous supposons que (Xk) est une marche aléatoire dont la loi des incréments est connue à un facteur d'échelle a près. Nous supposons que, pour tout k, Yk est une observation de f(Xk) dans un bruit additif gaussien, où f est une fonction que nous cherchons à estimer. Nous établissons l'identifiabilité du modèle statistique et nous proposons une estimation de f et de a à partir de la vraisemblance par paires des observations. / This document is dedicated to inference problems in hidden Markov models. The first part is devoted to an online maximum likelihood estimation procedure which does not store the observations. We propose a new Expectation Maximization based method called the Block Online Expectation Maximization (BOEM) algorithm. This algorithm solves the online estimation problem for general hidden Markov models. In complex situations, it requires the introduction of Sequential Monte Carlo methods to approximate several expectations under the fixed interval smoothing distributions. The convergence of the algorithm is shown under the assumption that the Lp mean error due to the Monte Carlo approximation can be controlled explicitly in the number of observations and in the number of particles. Therefore, a second part of the document establishes such controls for several Sequential Monte Carlo algorithms. This BOEM algorithm is then used to solve the simultaneous localization and mapping problem in different frameworks. Finally, the last part of this thesis is dedicated to nonparametric estimation in hidden Markov models. It is assumed that the Markov chain (Xk) is a random walk lying in a compact set with increment distribution known up to a scaling factor a. At each time step k, Yk is a noisy observations of f(Xk) where f is an unknown function. We establish the identifiability of the statistical model and we propose estimators of f and a based on the pairwise likelihood of the observations.