This work presents a novel approach for the design of a predictive model of music that can be used to analyze and generate musical material that is highly context dependent. The system is based on an approach known as n-gram modeling, often used in language processing and speech recognition algorithms, implemented initially upon a framework of Variable-Length Markov Models (VLMMs) and then extended to Variable-Length Hidden Markov Models (VLHMMs). The system brings together various principles like escape probabilities, smoothing schemes and uses multiple representations of the data stream to construct a multiple viewpoints system that enables it to draw complex relationships between the different input n-grams, and use this information to provide a stronger prediction scheme. It is implemented as a MAX/MSP external in C++ and is intended to be a predictive framework that can be used to create generative music systems and educational and compositional tools for music. A formal quantitative evaluation scheme based on entropy of the predictions is used to evaluate the model in sequence prediction tasks on a database of tabla compositions. The results show good model performance for both the VLMM and the VLHMM while highlighting the expensive computational cost of higher-order VLHMMs.

M. Tech. Electrical Engineering / Determines whether Hidden Markov models (HMM) can be used to classify steady state visual evoked electroencephalogram signals in a BCI system. This is for the purpose of aiding disabled people in driving a wheelchair.

In this thesis we develop a class of nonlinear generative models for high-dimensional time series. The first key property of these models is their distributed, or "componential" latent state, which is characterized by binary stochastic variables which interact to explain the data. The second key property is the use of an undirected graphical model to represent the relationship between latent state (features) and observations. The final key property is composability: the proposed class of models can form the building blocks of deep networks by successively training each model on the features extracted by the previous one. We first propose a model based on the Restricted Boltzmann Machine (RBM) that uses an undirected model with binary latent variables and real-valued "visible" variables. The latent and visible variables at each time step receive directed connections from the visible variables at the last few time-steps. This "conditional" RBM (CRBM) makes on-line inference efficient and allows us to use a simple approximate learning procedure. We demonstrate the power of our approach by synthesizing various motion sequences and by performing on-line filling in of data lost during motion capture. We also explore CRBMs as priors in the context of Bayesian filtering applied to multi-view and monocular 3D person tracking. We extend the CRBM in a way that preserves its most important computational properties and introduces multiplicative three-way interactions that allow the effective interaction weight between two variables to be modulated by the dynamic state of a third variable. We introduce a factoring of the implied three-way weight tensor to permit a more compact parameterization. The resulting model can capture diverse styles of motion with a single set of parameters, and the three-way interactions greatly improve its ability to blend motion styles or to transition smoothly among them. In separate but related work, we revisit Products of Hidden Markov Models (PoHMMs). We show how the partition function can be estimated reliably via Annealed Importance Sampling. This enables us to demonstrate that PoHMMs outperform various flavours of HMMs on a variety of tasks and metrics, including log likelihood.

No description available.

The rapid development of corpus-based speech systems such as concatenative synthesis systems for under-resourced languages requires an efficient, consistent and accurate solution with regard to phonetic speech segmentation. Manual development of phonetically annotated corpora is a time consuming and expensive process which suffers from challenges regarding consistency and reproducibility, while automation of this process has only been satisfactorily demonstrated on large corpora of a select few languages by employing techniques requiring extensive and specialised resources. In this work we considered the problem of phonetic segmentation in the context of developing small prototypical speech synthesis corpora for new under-resourced languages. This was done through an empirical evaluation of existing segmentation techniques on typical speech corpora in three South African languages. In this process, the performance of these techniques were characterised under different data conditions and the efficient application of these techniques were investigated in order to improve the accuracy of resulting phonetic alignments. We found that the application of baseline speaker-specific Hidden Markov Models results in relatively robust and accurate alignments even under extremely limited data conditions and demonstrated how such models can be developed and applied efficiently in this context. The result is segmentation of sufficient quality for synthesis applications, with the quality of alignments comparable to manual segmentation efforts in this context. Finally, possibilities for further automated refinement of phonetic alignments were investigated and an efficient corpus development strategy was proposed with suggestions for further work in this direction. / Thesis (M.Ing. (Computer Engineering))--North-West University, Potchefstroom Campus, 2009.

The document is my habilitation thesis, which is a prerequisite for obtaining the "habilitation à diriger des recherche (HDR)" in France (https://fr.wikipedia.org/wiki/Habilitation_universitaire#En_France). The thesis is of cumulative form, thus providing an overview of my published works until summer 2013.

The rapid development of corpus-based speech systems such as concatenative synthesis systems for under-resourced languages requires an efficient, consistent and accurate solution with regard to phonetic speech segmentation. Manual development of phonetically annotated corpora is a time consuming and expensive process which suffers from challenges regarding consistency and reproducibility, while automation of this process has only been satisfactorily demonstrated on large corpora of a select few languages by employing techniques requiring extensive and specialised resources. In this work we considered the problem of phonetic segmentation in the context of developing small prototypical speech synthesis corpora for new under-resourced languages. This was done through an empirical evaluation of existing segmentation techniques on typical speech corpora in three South African languages. In this process, the performance of these techniques were characterised under different data conditions and the efficient application of these techniques were investigated in order to improve the accuracy of resulting phonetic alignments. We found that the application of baseline speaker-specific Hidden Markov Models results in relatively robust and accurate alignments even under extremely limited data conditions and demonstrated how such models can be developed and applied efficiently in this context. The result is segmentation of sufficient quality for synthesis applications, with the quality of alignments comparable to manual segmentation efforts in this context. Finally, possibilities for further automated refinement of phonetic alignments were investigated and an efficient corpus development strategy was proposed with suggestions for further work in this direction. / Thesis (M.Ing. (Computer Engineering))--North-West University, Potchefstroom Campus, 2009.

In this thesis, we developed Bayesian models and machine learning algorithms for protein secondary structure and beta-sheet prediction problems. In protein secondary structure prediction, we developed hidden semi-Markov models, N-best algorithms and training set reduction procedures for proteins in the single-sequence category. We introduced three residue dependency models (both probabilistic and heuristic) incorporating the statistically significant amino acid correlation patterns at structural segment borders. We allowed dependencies to positions outside the segments to relax the condition of segment independence. Another novelty of the models is the dependency to downstream positions, which is important due to asymmetric correlation patterns observed uniformly in structural segments. Among the dataset reduction methods, we showed that the composition based reduction generated the most accurate results. To incorporate non-local interactions characteristic of beta-sheets, we developed two N-best algorithms and a Bayesian beta-sheet model. In beta-sheet prediction, we developed a Bayesian model to characterize the conformational organization of beta-sheets and efficient algorithms to compute the optimum architecture, which includes beta-strand pairings, interaction types (parallel or anti-parallel) and residue-residue interactions (contact maps). We introduced a Bayesian model for proteins with six or less beta-strands, in which we model the conformational features in a probabilistic framework by combining the amino acid pairing potentials with a priori knowledge of beta-strand organizations. To select the optimum beta-sheet architecture, we analyzed the space of possible conformations by efficient heuristics, in which we significantly reduce the search space by enforcing the amino acid pairs that have strong interaction potentials. For proteins with more than six beta-strands, we first computed beta-strand pairings using the BetaPro method. Then, we computed gapped alignments of the paired beta-strands in parallel and anti-parallel directions and chose the interaction types and beta-residue pairings with maximum alignment scores. Accurate prediction of secondary structure, beta-sheets and non-local contacts should improve the accuracy and quality of the three-dimensional structure prediction.

In this study, an ensemble speaker and speaking environment modeling (ESSEM) approach is proposed to characterize environments in order to enhance performance robustness of automatic speech recognition (ASR) systems under adverse conditions. The ESSEM process comprises two stages, the offline and online phases. In the offline phase, we prepare an ensemble speaker and speaking environment space formed by a collection of super-vectors. Each super-vector consists of the entire set of means from all the Gaussian mixture components of a set of hidden Markov Models that characterizes a particular environment. In the online phase, with the ensemble environment space prepared in the offline phase, we estimate the super-vector for a new testing environment based on a stochastic matching criterion. A series of techniques is proposed to further improve the original ESSEM approach on both offline and online phases. For the offline phase, we focus on methods to enhance the construction and coverage of the environment space. We first demonstrate environment clustering and environment partitioning algorithms to well structure the environment space; then, we propose a discriminative training algorithm to enhance discrimination across environment super-vectors and therefore broaden the coverage of the ensemble environment space. For the online phase, we study methods to increase the efficiency and precision in estimating the target super-vector for the testing condition. To enhance the efficiency, we incorporate dimensionality reduction techniques to reduce the complexity of the original environment space. To improve the precision, we first study different forms of mapping function and propose a weighted N-best information technique; then, we propose cohort selection, environment space adaptation and multiple cluster matching algorithms to facilitate the environment characterization. We evaluate the proposed ESSEM framework on the Aurora-2 connected digit recognition task. Experimental results verify that the original ESSEM approach already provides clear improvement over a baseline system without environment compensation. Moreover, the performance of ESSEM can be further enhanced by using the proposed offline and online algorithms. A significant improvement of 16.08% word error rate reduction is achieved by ESSEM with optimal offline and online configuration over our best baseline system on the Aurora-2 task.

This thesis describes new gene finding methods for eukaryotic gene prediction. The current methods for deriving model parameters for gene prediction algorithms are based on curated or experimentally validated set of genes or gene elements. These training sets often require time and additional expert efforts especially for the species that are in the initial stages of genome sequencing. Unsupervised training allows determination of model parameters from anonymous genomic sequence with. The importance and the practical applicability of the unsupervised training is critical for ever growing rate of eukaryotic genome sequencing. Three distinct training procedures are developed for diverse group of eukaryotic species. GeneMark-ES is developed for species with strong donor and acceptor site signals such as Arabidopsis thaliana, Caenorhabditis elegans and Drosophila melanogaster. The second version of the algorithm, GeneMark-ES-2, introduces enhanced intron model to better describe the gene structure of fungal species with posses with relatively weak donor and acceptor splice sites and well conserved branch point signal. GeneMark-LE, semi-supervised training approach is designed for eukaryotic species with small number of introns. The results indicate that the developed unsupervised training methods perform well as compared to other training methods and as estimated from the set of genes supported by EST-to-genome alignments. Analysis of novel genomes reveals interesting biological findings and show that several candidates of under-annotated and over-annotated fungal species are present in the current set of annotated of fungal genomes.