Multi-Pattern Viterbi Algorithm for joint decoding of multiple speech patterns

Joint decoding of multiple speech patterns so as to improve speech recognition performance is important, especially in the presence of noise. In this paper, we propose a Multi-Pattern Viterbi algorithm (MPVA) to jointly decode and recognize multiple speech patterns for automatic speech recognition (ASR). The MPVA is a generalization of the Viterbi Algorithm to jointly decode multiple patterns given a Hidden Markov Model (HMM). Unlike the previously proposed two stage Constrained Multi-Pattern Viterbi Algorithm (CMPVA),the MPVA is a single stage algorithm. MPVA has the advantage that it cart be extended to connected word recognition (CWR) and continuous speech recognition (CSR) problems. MPVA is shown to provide better speech recognition performance than the earlier techniques: using only two repetitions of noisy speech patterns (-5 dB SNR, 10% burst noise), the word error rate using MPVA decreased by 28.5%, when compared to using individual decoding. (C) 2010 Elsevier B.V. All rights reserved.

Simulation-based optimal sensor scheduling with application to observer trajectory planning

The sensor scheduling problem can be formulated as a controlled hidden Markov model and this paper solves the problem when the state, observation and action spaces are continuous. This general case is important as it is the natural framework for many applications. The aim is to minimise the variance of the estimation error of the hidden state w.r.t. the action sequence. We present a novel simulation-based method that uses a stochastic gradient algorithm to find optimal actions. © 2007 Elsevier Ltd. All rights reserved.

Filtering via approximate Bayesian computation

Approximate Bayesian computation (ABC) has become a popular technique to facilitate Bayesian inference from complex models. In this article we present an ABC approximation designed to perform biased filtering for a Hidden Markov Model when the likelihood function is intractable. We use a sequential Monte Carlo (SMC) algorithm to both fit and sample from our ABC approximation of the target probability density. This approach is shown to, empirically, be more accurate w.r.t.~the original filter than competing methods. The theoretical bias of our method is investigated; it is shown that the bias goes to zero at the expense of increased computational effort. Our approach is illustrated on a constrained sequential lasso for portfolio allocation to 15 constituents of the FTSE 100 share index.

Sequential decision making in general state space models

Many problems in control and signal processing can be formulated as sequential decision problems for general state space models. However, except for some simple models one cannot obtain analytical solutions and has to resort to approximation. In this thesis, we have investigated problems where Sequential Monte Carlo (SMC) methods can be combined with a gradient based search to provide solutions to online optimisation problems. We summarise the main contributions of the thesis as follows. Chapter 4 focuses on solving the sensor scheduling problem when cast as a controlled Hidden Markov Model. We consider the case in which the state, observation and action spaces are continuous. This general case is important as it is the natural framework for many applications. In sensor scheduling, our aim is to minimise the variance of the estimation error of the hidden state with respect to the action sequence. We present a novel SMC method that uses a stochastic gradient algorithm to find optimal actions. This is in contrast to existing works in the literature that only solve approximations to the original problem. In Chapter 5 we presented how an SMC can be used to solve a risk sensitive control problem. We adopt the use of the Feynman-Kac representation of a controlled Markov chain flow and exploit the properties of the logarithmic Lyapunov exponent, which lead to a policy gradient solution for the parameterised problem. The resulting SMC algorithm follows a similar structure with the Recursive Maximum Likelihood(RML) algorithm for online parameter estimation. In Chapters 6, 7 and 8, dynamic Graphical models were combined with with state space models for the purpose of online decentralised inference. We have concentrated more on the distributed parameter estimation problem using two Maximum Likelihood techniques, namely Recursive Maximum Likelihood (RML) and Expectation Maximization (EM). The resulting algorithms can be interpreted as an extension of the Belief Propagation (BP) algorithm to compute likelihood gradients. In order to design an SMC algorithm, in Chapter 8 uses a nonparametric approximations for Belief Propagation. The algorithms were successfully applied to solve the sensor localisation problem for sensor networks of small and medium size.

Cloning and expression of zebra fish ferroportin and investigating its influence on anemia

Iron is required for many microbes and pathogens for their survival and proliferation including Leishmania which cause leishmaniasis. Leishmaniasis is an increasingly serious infectious disease with a wide spectrum of clinical manifestations. These range from localized cutaneous leishmaniasis (CL) lesions to a lethal visceral form. Certain strains such as BALB/c mice fail to control L. major infection and develop progressive lesions and systemic disease. These mice are thought to be a model of non-healing forms of the human disease such as kala-azar or diffuse cutaneous leishmaniasis. Progression of disease in BALB/c mice has been associated with the anemia, in last days of their survival, the progressive anemia is considered to be one of the reasons of their death. Ferroportin (Fpn), a key regulator of iron homeostasis is a conserved membrane protein that exports iron across the duodenal enterocytes as well as macrophages and hepatocytes into the blood circulation. Fpn has also critical influence on survival and proliferation of many microorganisms whose growth is dependent upon iron, thus preparation of Fpn is needed to study the role of iron in immune responses and pathogenesis of micoorganisms. To prepare and characterize a recombinant ferroportin, total RNA was extracted from Indian zebrafish duodenum, and used to synthesize cDNA by RT-PCR. PCR product was first cloned in Topo TA vector and then subcloned into the GFP expression vector pEGFP–N1. The final resulted plasmid (pEGFP-ZFpn) was used for expression of FPN-EGFP protein in Hek 293T cells. The expression was confirmed by fluorescence microscopy and flow cytometery. Recombinant Fpn was further characterized by submission of its predicted amino acid sequences to the TMHMM V2.0 prediction server (hidden Markov model), NetOGlyc 3.1 server and NetNGlyc 3.1 server. Data emphasised that obtained Fpn from indian zebrafish contained eight transmembrane domains with N- and C-termini inside the cytoplasm and harboured 78 mucin-type glycosylated amino acid. The results indicate that the prepared and characterized recombinant Fpn protein has no membrane topology difference compared to other Fpn described by other researcher. Our next aim was to deliver recombinant plasmid (pEGFP-ZFpn) to entrocyte cells. However, naked therapeutic genes are rapidly degraded by nucleases, showing poor cellular uptake, nonspecificity to the target cells, and low transfection efficiency. The development of safe and efficient gene carriers is one of the prerequisites for the success of gene therapy. Chitosan and alginate 139 polymers were used for oral gene carrier because of their biodegradability, biocompatibility and their mucoadhesive and permeability-enhancing properties in the gut. Nanoparticles comprising Alginate/Chitosan polymers were prepared by pregel preparation method. The resulting nanoparticles had a loading efficiency of 95% and average size of 188 nm as confirmed by PCS method and SEM images had showed spherical particles. BALB/c mice were divided to three groups. The first and second group were fed with chitosan/alginate nanoparticles containing the pEGFP-ZFpn and pEGFP plasmid, respectively (30 μgr/mice) and the third group (control) didn’t get any nanoparticles. The result showed BALB/c mice infected by L.major, resulted in higher hematocryte and iron level in pEGFP-ZFpn fed mice than that in other groups. Consentration of cytokines determined by ELISA showed lower levels of IL-4 and IL-10 and higher levels of IFN-γ/IL-4 and IFN-γ/IL-10 ratios in pEGFP-ZFpn fed mice than that in other groups. Morover more limited increase of footpad thickness and significant reduction of viable parasites in lymph node was seen in pEGFP-ZFpn fed mice. The results showed the first group exhibited a highr hematocryte and iron compared to the other groups. These data strongly suggests the in vivo administration of chitosan/alginate nanoparticles containing pEGFP-ZFpn suppress Th2 response and may be used to control the leishmaniasis .

Unsupervised intra-lingual and cross-lingual speaker adaptation for HMM-based speech synthesis using two-pass decision tree construction

Hidden Markov model (HMM)-based speech synthesis systems possess several advantages over concatenative synthesis systems. One such advantage is the relative ease with which HMM-based systems are adapted to speakers not present in the training dataset. Speaker adaptation methods used in the field of HMM-based automatic speech recognition (ASR) are adopted for this task. In the case of unsupervised speaker adaptation, previous work has used a supplementary set of acoustic models to estimate the transcription of the adaptation data. This paper first presents an approach to the unsupervised speaker adaptation task for HMM-based speech synthesis models which avoids the need for such supplementary acoustic models. This is achieved by defining a mapping between HMM-based synthesis models and ASR-style models, via a two-pass decision tree construction process. Second, it is shown that this mapping also enables unsupervised adaptation of HMM-based speech synthesis models without the need to perform linguistic analysis of the estimated transcription of the adaptation data. Third, this paper demonstrates how this technique lends itself to the task of unsupervised cross-lingual adaptation of HMM-based speech synthesis models, and explains the advantages of such an approach. Finally, listener evaluations reveal that the proposed unsupervised adaptation methods deliver performance approaching that of supervised adaptation.

Robust estimation of local genetic ancestry in admixed populations using a nonparametric Bayesian approach.

We present a new haplotype-based approach for inferring local genetic ancestry of individuals in an admixed population. Most existing approaches for local ancestry estimation ignore the latent genetic relatedness between ancestral populations and treat them as independent. In this article, we exploit such information by building an inheritance model that describes both the ancestral populations and the admixed population jointly in a unified framework. Based on an assumption that the common hypothetical founder haplotypes give rise to both the ancestral and the admixed population haplotypes, we employ an infinite hidden Markov model to characterize each ancestral population and further extend it to generate the admixed population. Through an effective utilization of the population structural information under a principled nonparametric Bayesian framework, the resulting model is significantly less sensitive to the choice and the amount of training data for ancestral populations than state-of-the-art algorithms. We also improve the robustness under deviation from common modeling assumptions by incorporating population-specific scale parameters that allow variable recombination rates in different populations. Our method is applicable to an admixed population from an arbitrary number of ancestral populations and also performs competitively in terms of spurious ancestry proportions under a general multiway admixture assumption. We validate the proposed method by simulation under various admixing scenarios and present empirical analysis results from a worldwide-distributed dataset from the Human Genome Diversity Project.

Inter-Frame Contextual Modelling For Visual Speech Recognition

In this paper, we present a new approach to visual speech recognition which improves contextual modelling by combining Inter-Frame Dependent and Hidden Markov Models. This approach captures contextual information in visual speech that may be lost using a Hidden Markov Model alone. We apply contextual modelling to a large speaker independent isolated digit recognition task, and compare our approach to two commonly adopted feature based techniques for incorporating speech dynamics. Results are presented from baseline feature based systems and the combined modelling technique. We illustrate that both of these techniques achieve similar levels of performance when used independently. However significant improvements in performance can be achieved through a combination of the two. In particular we report an improvement in excess of 17% relative Word Error Rate in comparison to our best baseline system.

How to estimate scavenger fish abundance using baited camera data

Baited cameras are often used for abundance estimation wherever alternative techniques are precluded, e.g. in abyssal systems and areas such as reefs. This method has thus far used models of the arrival process that are deterministic and, therefore, permit no estimate of precision.
Furthermore, errors due to multiple counting of fish and missing those not seen by the camera have restricted the technique to using only the time of first arrival, leaving a lot of data redundant. Here, we reformulate the arrival process using a stochastic model, which allows the precision of abundance
estimates to be quantified. Assuming a non-gregarious, cross-current-scavenging fish, we show that prediction of abundance from first arrival time is extremely uncertain. Using example data, we show
that simple regression-based prediction from the initial (rising) slope of numbers at the bait gives good precision, accepting certain assumptions. The most precise abundance estimates were obtained
by including the declining phase of the time series, using a simple model of departures, and taking account of scavengers beyond the camera’s view, using a hidden Markov model.

Probabilistic Inference in Credal Networks: New Complexity Results

Credal networks are graph-based statistical models whose parameters take values in a set, instead of being sharply specified as in traditional statistical models (e.g., Bayesian networks). The computational complexity of inferences on such models depends on the irrelevance/independence concept adopted. In this paper, we study inferential complexity under the concepts of epistemic irrelevance and strong independence. We show that inferences under strong independence are NP-hard even in trees with binary variables except for a single ternary one. We prove that under epistemic irrelevance the polynomial-time complexity of inferences in credal trees is not likely to extend to more general models (e.g., singly connected topologies). These results clearly distinguish networks that admit efficient inferences and those where inferences are most likely hard, and settle several open questions regarding their computational complexity. We show that these results remain valid even if we disallow the use of zero probabilities. We also show that the computation of bounds on the probability of the future state in a hidden Markov model is the same whether we assume epistemic irrelevance or strong independence, and we prove an analogous result for inference in Naive Bayes structures. These inferential equivalences are important for practitioners, as hidden Markov models and Naive Bayes networks are used in real applications of imprecise probability.

An HMM-based spectrum occupancy predictor for energy efficient cognitive radio

Spectrum sensing is the cornerstone of cognitive radio technology and refers to the process of obtaining awareness of the radio spectrum usage in order to detect the presence of other users. Spectrum sensing algorithms consume considerable energy and time. Prediction methods for inferring the channel occupancy of future time instants have been proposed as a means of improving performance in terms of energy and time consumption. This paper studies the performance of a hidden Markov model (HMM) spectrum occupancy predictor as well as the improvement in sensing energy and time consumption based on real occupancy data obtained in the 2.4GHz ISM band. Experimental results show that the HMM-based occupancy predictor outperforms a kth order Markov and a 1-nearest neighbour (1NN) predictor. Our study also suggests that by employing such a predictive scheme in spectrum sensing, an improvement of up to 66% can be achieved in the required sensing energy and time.

Adaptive filtering for high quality HMM based speech synthesis

In this work an adaptive filtering scheme based on a dual Discrete Kalman Filtering (DKF) is proposed for Hidden Markov Model (HMM) based speech synthesis quality enhancement. The objective is to improve signal smoothness across HMMs and their related states and to reduce artifacts due to acoustic model's limitations. Both speech and artifacts are modelled by an autoregressive structure which provides an underlying time frame dependency and improves time-frequency resolution. Themodel parameters are arranged to obtain a combined state-space model and are also used to calculate instantaneous power spectral density estimates. The quality enhancement is performed by a dual discrete Kalman filter that simultaneously gives estimates for the models and the signals. The system's performance has been evaluated using mean opinion score tests and the proposed technique has led to improved results.

Speech Recognition of Malayalam Numbers

Digit speech recognition is important in many applications such as automatic data entry, PIN entry, voice dialing telephone, automated banking system, etc. This paper presents speaker independent speech recognition system for Malayalam digits. The system employs Mel frequency cepstrum coefficient (MFCC) as feature for signal processing and Hidden Markov model (HMM) for recognition. The system is trained with 21 male and female voices in the age group of 20 to 40 years and there was 98.5% word recognition accuracy (94.8% sentence recognition accuracy) on a test set of continuous digit recognition task.

Automated Transcription System for Malayalam Language

Malayalam is one of the 22 scheduled languages in India with more than 130 million speakers. This paper presents a report on the development of a speaker independent, continuous transcription system for Malayalam. The system employs Hidden Markov Model (HMM) for acoustic modeling and Mel Frequency Cepstral Coefficient (MFCC) for feature extraction. It is trained with 21 male and female speakers in the age group ranging from 20 to 40 years. The system obtained a word recognition accuracy of 87.4% and a sentence recognition accuracy of 84%, when tested with a set of continuous speech data.

Malayalam Isolated Digit Recognition using HMM and PLP cepstral coefficient

Development of Malayalam speech recognition system is in its infancy stage; although many works have been done in other Indian languages. In this paper we present the first work on speaker independent Malayalam isolated speech recognizer based on PLP (Perceptual Linear Predictive) Cepstral Coefficient and Hidden Markov Model (HMM). The performance of the developed system has been evaluated with different number of states of HMM (Hidden Markov Model). The system is trained with 21 male and female speakers in the age group ranging from 19 to 41 years. The system obtained an accuracy of 99.5% with the unseen data