Discrete Hidden Markov Models Research Articles

Speech command recognition using a system based on the digital signal processor TMS320F2812 of Texas Instruments.

This paper describes the hardware design of a multi‐purpose electronic system for speech command recognition based on the digital signal processor TMS320F2812 of Texas Instruments. In the paper, we utilize this system for speech command recognition. Specifically, we design and build an electronic system capable of recognizing speech commands. The main blocks are a power source, oscillator, adc, amplifier, jtag, reset boot mode, and monitor LED. The technique used for recognition is a multi section vector quantization, which is a well known technique in the field. This was selected over techniques such as discrete hidden Markov models (HMMs), Gaussian mixture models, or dynamic time warping. It is faster and equally precise compared to these techniques for the purpose of a very limited vocabulary, although we are aware that several types of HMMs are the techniques more widely used until now for continuous speech recognition systems. Preliminary tests for five words give us a rate of 96% in recognition accuracy for 10 repetitions for training and 10 for testing (one subject, under a 20 dB of signal‐to‐noise ratio). In these months, the system will be tested for more words and subjects. Also, it will be tested in real situations.

Biomedical system based on the Discrete Hidden Markov Model using the Rocchio–Genetic approach for the classification of internal carotid artery Doppler signals

When the maximum likelihood approach (ML) is used during the calculation of the Discrete Hidden Markov Model (DHMM) parameters, DHMM parameters of the each class are only calculated using the training samples (positive training samples) of the same class. The training samples (negative training samples) not belonging to that class are not used in the calculation of DHMM model parameters. With the aim of supplying that deficiency, by involving the training samples of all classes in calculating processes, a Rocchio algorithm based approach is suggested. During the calculation period, in order to determine the most appropriate values of parameters for adjusting the relative effect of the positive and negative training samples, a Genetic algorithm is used as an optimization technique. The purposed method is used to classify the internal carotid artery Doppler signals recorded from 136 patients as well as of 55 healthy people. Our proposed method reached 97.38% classification accuracy with fivefold cross-validation (CV) technique. The classification results showed that the proposed method was effective for the classification of internal carotid artery Doppler signals.

English

There are two paradigms for modelling the varying length temporal data namely, modelling the sequences of feature vectors as in the hidden Markov model-based approaches for speech recognition and modelling the sets of feature vectors as in the Gaussian mixture model (GMM)-based approaches for speech emotion recognition. In this paper, the methods using discrete hidden Markov models (DHMMs) in the kernel feature space and string kernel-based SVM classifier for classification of discretised representation of sequence of feature vectors obtained by clustering and vector quantisation in the kernel feature space are presented. The authors then present continuous density hidden Markov models (CDHMMs) in the explicit kernel feature space that use the continuous valued representation of features extracted from the temporal data. The methods for temporal pattern classification by mapping a varying length sequential pattern to a fixed-length sequential pattern and then using an SVM-based classifier for classification are also presented. The task of recognition of spoken letters in E-set, it is possible to build models that use a discretised representation and string kernel SVM based classification and obtain a classification performance better than that of models using the continuous valued representation is demonstrated. For modelling sets of vectors-based representation of temporal data, two approaches in a hybrid framework namely, the score vector-based approach and the segment modelling based approach are presented. In both approaches, a generative model-based method is used to obtain a fixed length pattern representation for a varying length temporal data and then a discriminative model is used for classification. These two approaches are studied for speech emotion recognition task. The segment modelling based approach gives a better performance than the score vector-based approach and the GMM-based classifiers for speech emotion recognition.Defence Science Journal, 2010, 60(4), pp.348-363, DOI:http://dx.doi.org/10.14429/dsj.60.492

An efficient speech recognition system in adverse conditions using the nonparametric regression

General Regression Neural Networks (GRNN) have been applied to phoneme identification and isolated word recognition in clean speech. In this paper, the authors extended this approach to Arabic spoken word recognition in adverse conditions. In fact, noise robustness is one of the most challenging problems in Automatic Speech Recognition (ASR) and most of the existing recognition methods, which have shown to be highly efficient under noise-free conditions, fail drastically in noisy environments. The proposed system was tested for Arabic digit recognition at different Signal-to-Noise Ratio (SNR) levels and under four noisy conditions: multispeakers babble background, car production hall (factory), military vehicle (leopard tank) and fighter jet cockpit (buccaneer) issued from NOISEX-92 database. The proposed scheme was successfully compared to the similar recognizers based on the Multilayer Perceptrons (MLP), the Elman Recurrent Neural Network (RNN) and the discrete Hidden Markov Model (HMM). The experimental results showed that the use of nonparametric regression with an appropriate smoothing factor (spread) improved the generalization power of the neural network and the global performance of the speech recognizer in noisy environments.

A new approach based on discrete hidden Markov model using Rocchio algorithm for the diagnosis of the brain diseases

Transcranial Doppler (TCD) study of the adult intracerebral circulation has gained an important popularity in last 10 years, since it is a non-invasive, easy to apply and reliable technique. In this study, an implementation on biomedical system has been developed for classification of signals gathered from middle cerebral arteries in the temporal area via TCD for 24 healthy and 82 ill people which have one of the four different brain patients such as; cerebral aneurysm, brain hemorrhage, cerebral oedema and brain tumor. Basically, the system is composed of feature extraction and classification parts. In the feature extraction stage, the Linear Predictive Coding (LPC) Analysis and Cepstral Analysis were applied in order to extract the cepstral and delta-cepstral coefficients in frame level as feature vectors. In the classification stage a new Discrete Hidden Markov Model (DHMM) based approach was proposed for the diagnosis of brain diseases. This proposed method was developed via Rocchio algorithm. Therefore, to calculate DHMM parameters regulated according to maximum likelihood (ML) approach, both training samples of related class and other classes were included in calculation. Thus, DHMM model parameters presenting one class were suggested to represent the training samples related to that class better as well as not to represent the training samples related to other classes. The performance of the proposed DHMM with Rocchio approach was compared with some methods such as DHMM, Artificial Neural Network (ANN), neuro-fuzzy approaches and obtained better classification performance than these methods.

An enhanced independent component-based human facial expression recognition from video

Facial expression recognition (FER) from video is an essential research area in the field of human computer interfaces (HCI). In this work, we present a new method to recognize several facial expressions from time sequential facial expression images. To produce robust facial expression features, enhanced independent component analysis (EICA) is utilized to extract locally independent component (IC) features which are further classified by Fisher linear discriminant analysis (FLDA). Using these features, discrete hidden Markov models (HMMs) are utilized to model different facial expressions such as joy, anger, and sad. Performance of our proposed FER system is compared against four other conventional feature extraction approaches (i.e., PCA, PCA-FLDA, ICA, and EICA) in conjunction with the same HMM scheme. The experimental results using the Cohn-Kanade database of facial expression videos show that our proposed system yields much improved recognition rate reaching the mean recognition rate of 93.23% whereas the conventional methods yield 82.92% at best.

Non-Homogeneous Multivariate Integrable HMM

This paper derives the PDF and CDF for a diffusion process controlled by a discrete time multi-variate non-homogeneous HMM. This method has multiple applications from options pricing to market making. The chief advantages include descrete regime switching ability, non-Gaussian, nonconstant variance noise, the use of exogeneous covariates and reasonably low computational complexity.

Improvement of the speech recognition in noisy environments using a nonparametric regression

In this paper, an efficient speech recognition system based on the general regression neural network (GRNN) has been presented. The GRNN has been previously applied for phoneme identification and isolated word recognition in quiet environment. We propose to extend this method to Arabic spoken word recognition in adverse conditions because noise robustness is one of the most challenging problems in automatic speech recognition (ASR). The proposed system has been tested for Arabic digit recognition at different signal-to-noise ratio (SNR) levels in various noisy conditions, including stationary and nonstationary background noises issued from NOISEX-92 database. The proposed scheme is compared with the similar recognisers based on the multilayer perceptron (MLP), the Elman recurrent neural network (RNN) and the discrete hidden Markov model (HMM). The experimental results show that the use of the neural network approach including nonparametric regression improves the global performance of the speech recogniser in noisy environments.

On the Use of Hidden Markov Modeling and Time-frequency Features for Damage Classification in Composite Structures

A novel approach based on hidden Markov models (HMMs) is proposed for damage classification in composite structures. Time-frequency damage features are first extracted from the measured signals using the matching pursuit decomposition algorithm. The features are then incorporated as observation sequences to be modeled statistically by the HMMs. Once built, the HMMs are integrated very efficiently into a Bayesian framework for the classification of structural damage. Both discrete and continuous observation density HMMs are considered; continuous HMMs are shown to yield better accuracy, but at the cost of added computational complexity. A decision fusion procedure is employed to combine the local classification results at each sensor, significantly enhancing the overall classification performance. The utility of the proposed technique is demonstrated by its application to the classification of delamination damage, impact damage, and progressive tensile damage in laminated composites.

Discriminative Training of the Hidden Vector State Model for Semantic Parsing

In this paper, we discuss how discriminative training can be applied to the hidden vector state (HVS) model in different task domains. The HVS model is a discrete hidden Markov model (HMM) in which each HMM state represents the state of a push-down automaton with a finite stack size. In previous applications, maximum-likelihood estimation (MLE) is used to derive the parameters of the HVS model. However, MLE makes a number of assumptions and unfortunately some of these assumptions do not hold. Discriminative training, without making such assumptions, can improve the performance of the HVS model by discriminating the correct hypothesis from the competing hypotheses. Experiments have been conducted in two domains: the travel domain for the semantic parsing task using the DARPA Communicator data and the Air Travel Information Services (ATIS) data and the bioinformatics domain for the information extraction task using the GENIA corpus. The results demonstrate modest improvements of the performance of the HVS model using discriminative training. In the travel domain, discriminative training of the HVS model gives a relative error reduction rate of 31 percent in F-measure when compared with MLE on the DARPA Communicator data and 9 percent on the ATIS data. In the bioinformatics domain, a relative error reduction rate of 4 percent in F-measure is achieved on the GENIA corpus.

A new approach based on a discrete hidden Markov model using the Rocchio algorithm for the diagnosis of heart valve diseases

Abstract: Application of the Doppler ultrasound technique in the diagnosis of heart diseases has been increasing in the last decade since it is non‐invasive, practicable and reliable. In this study, a new approach based on the discrete hidden Markov model (DHMM) is proposed for the diagnosis of heart valve disorders. For the calculation of hidden Markov model (HMM) parameters according to the maximum likelihood approach, HMM parameters belonging to each class are calculated by using training samples that only belong to their own classes. In order to calculate the parameters of DHMMs, not only training samples of the related class but also training samples of other classes are included in the calculation. Therefore HMM parameters that reflect a class's characteristics are more represented than other class parameters. For this aim, the approach was to use a hybrid method by adapting the Rocchio algorithm. The proposed system was used in the classification of the Doppler signals obtained from aortic and mitral heart valves of 215 subjects. The performance of this classification approach was compared with the classification performances in previous studies which used the same data set and the efficiency of the new approach was tested. The total classification accuracy of the proposed approach (95.12%) is higher than the total accuracy rate of standard DHMM (94.31%), continuous HMM (93.5%) and support vector machine (92.67%) classifiers employed in our previous studies and comparable with the performance levels of classifications using artificial neural networks (95.12%) and fuzzy‐C‐means/CHMM (95.12%).

Continuous earthquake detection and classification using discrete Hidden Markov Models

SUMMARY We present a novel technique to solve the automatic detection and classification problem of earth tremor in a single step by using Hidden Markov Modelling (HMM). While this technique was originally developed in speech recognition, it already showed great promise when applied to volcano induced seismic signals. We apply the HMM classifier to a much simpler problem, that is, the detection and distance dependent classification of small to medium sized earthquakes. Using the smaller and possibly not perfect data set of earthquakes recorded with three stations of the Bavarian Earthquake Service enables us to better evaluate the advantages and disadvantages of the proposed algorithm and to compare the results with simple and widely used detection techniques (e.g. recursive short-term versus long-term average). Overall the performance of HMM shows good results in the pre-triggered classification tasks and reasonable results in the continuous case. The application of HMMs is illustrated step by step so it can be used as recipe for other applications. Special emphasize is given to the important problem of selecting the features, which best describe the properties of the different signals that are to be classified.

Off-line cursive handwritten Tamil character recognition

In spite of several advancements in technologies pertaining to Optical character recognition, handwriting continues to persist as means of documenting information for day-to-day life. The process of segmentation and recognition pose quiets a lot of challenges especially in recognizing cursive handwritten scripts of different languages. The concept proposed is a solution crafted to perform character recognition of hand-written scripts in Tamil, a language having official status in India, Sri Lanka, and Singapore. The approach utilizes discrete Hidden Markov Models (HMMs) for recognizing off-line cursive handwritten Tamil characters. The tolerance of the system is evident as it can overwhelm the complexities arise out of font variations and proves to be flexible and robust. Higher degree of accuracy in results has been obtained with the implementation of this approach on a comprehensive database and the precision of the results demonstrates its application on commercial usage. The methodology promises to present a simple and fast scaffold to construct a full OCR system extended with suitable pre-processing.

Recognising online spatial activities using a bioinformatics inspired sequence alignment approach

In this paper we address the problem of recognising embedded activities within continuous spatial sequences obtained from an online video tracking system. Traditionally, continuous data streams such as video tracking data are buffered with a sliding window applied to the buffered data stream for activity detection. We introduce an algorithm based on Smith–Waterman (SW) local alignment from the field of bioinformatics that can locate and accurately quantify embedded activities within a windowed sequence. The modified SW approach utilises dynamic programming with two dimensional spatial data to quantify sequence similarity and is capable of recognising sequences containing gaps and significant amounts of noise. A more efficient SW formulation for online recognition, called Online SW (OSW), is also developed. Through experimentation we show that the OSW algorithm can accurately and robustly recognise manually segmented activity sequences as well as embedded sequences from an online tracking system. To benchmark the classification performance of OSW we compare the approach to dynamic time warping (DTW) and the discrete hidden Markov model (HMM). Results demonstrate that OSW produces higher precision and recall than both DTW and the HMM in an online recognition context. With accurately segmented sequences the SW approach produces results comparable to DTW and superior to the HMM. Finally, we confirm the robust property of the SW approach by evaluating it with sequences containing artificially incorporated noise.

Human behavior classification from MPEG compressed videos

A new approach is proposed for human behavior classification from MPEG compressed videos. Moving objects are first detected by subtracting the dc values in I frames from those in the background. In addition, the dc values of the background are also adapted to avoid noise and illumination change. The tracking process is then performed using the consecutive frames. Motion vectors extracted from P frames are used to predict the next position of moving objects. An overlapping table is constructed to determine relationships between moving objects, and the number of moving objects is updated. For analyzing human behavior, motion vectors and velocities of moving objects from P and B frames are extracted. These features are clustered to codewords using a codebook generated by vector quantization (VQ) for the input of discrete hidden Markov models (HMMs). By applying the HMM, four kinds of human behaviors are successfully identified from the human behavior sequences. The proposed approach is, furthermore, more accurate than the previous method based on conventional features.

Integrated Detection, Tracking, and Recognition of Faces with Omnivideo Array in Intelligent Environments

We present a multilevel system architecture for intelligent environments equipped with omnivideo arrays. In order to gain unobtrusive human awareness, real-time 3D human tracking as well as robust video-based face detection and tracking and face recognition algorithms are needed. We first propose a multiprimitive face detection and tracking loop to crop face videos as the front end of our face recognition algorithm. Both skin-tone and elliptical detections are used for robust face searching, and view-based face classification is applied to the candidates before updating the Kalman filters for face tracking. For video-based face recognition, we propose three decision rules on the facial video segments. The majority rule and discrete HMM (DHMM) rule accumulate single-frame face recognition results, while continuous density HMM (CDHMM) works directly with the PCA facial features of the video segment for accumulated maximum likelihood (ML) decision. The experiments demonstrate the robustness of the proposed face detection and tracking scheme and the three streaming face recognition schemes with 99% accuracy of the CDHMM rule. We then experiment on the system interactions with single person and group people by the integrated layers of activity awareness.We also discuss the speech-aided incremental learning of new faces.

Ensemble of HMM classifiers based on the clustering validity index for a handwritten numeral recognizer

A new scheme for the optimization of codebook sizes for Hidden Markov Models (HMMs) and the generation of HMM ensembles is proposed in this paper. In a discrete HMM, the vector quantization procedure and the generated codebook are associated with performance degradation. By using a selected clustering validity index, we show that the optimization of HMM codebook size can be selected without training HMM classifiers. Moreover, the proposed scheme yields multiple optimized HMM classifiers, and each individual HMM is based on a different codebook size. By using these to construct an ensemble of HMM classifiers, this scheme can compensate for the degradation of a discrete HMM.

Arabic Handwritten Word Recognition Using HMMs with Explicit State Duration

We describe an offline unconstrained Arabic handwritten word recognition system based on segmentation-free approach and discrete hidden Markov models (HMMs) with explicit state duration. Character durations play a significant part in the recognition of cursive handwriting. The duration information is still mostly disregarded in HMM-based automatic cursive handwriting recognizers due to the fact that HMMs are deficient in modeling character durations properly. We will show experimentally that explicit state duration modeling in the HMM framework can significantly improve the discriminating capacity of the HMMs to deal with very difficult pattern recognition tasks such as unconstrained Arabic handwriting recognition. In order to carry out the letter and word model training and recognition more efficiently, we propose a new version of the Viterbi algorithm taking into account explicit state duration modeling. Three distributions (Gamma, Gauss, and Poisson) for the explicit state duration modeling have been used, and a comparison between them has been reported. To perform word recognition, the described system uses an original sliding window approach based on vertical projection histogram analysis of the word and extracts a new pertinent set of statistical and structural features from the word image. Several experiments have been performed using the IFN/ENIT benchmark database and the best recognition performances achieved by our system outperform those reported recently on the same database.

Bayesian online algorithms for learning in discrete hidden Markov models

We propose and analyze two different Bayesian online algorithms for learning in discrete Hidden Markov Models and compare their performance with the already known Baldi-Chauvin Algorithm. Using the Kullback-Leibler divergence as a measure of generalization we draw learning curves in simplified situations for these algorithms and compare their performances.

A biomedical system based on fuzzy discrete hidden Markov model for the diagnosis of the brain diseases

Because it is a non-invasive, easy to apply and reliable technique, transcranial doppler (TCD) study of the adult intracerebral circulation has increased enormously in the last 10 years. In this study, a biomedical system has been implemented in order to classify the TCD signals recorded from the temporal region of the brain of 82 patients as well as of 24 healthy people. The diseases were investigated cerebral aneurysm, brain hemorrhage, cerebral oedema and brain tumor. The system is composed of feature extraction and classification parts, basically. In the feature extraction stage, the linear predictive coding analysis and cepstral analysis were applied in order to extract the cepstral and delta-cepstral coefficients in frame level as feature vectors. In the classification stage, discrete hidden Markov model (DHMM) based methods were used. In order to avoid loosing information due to vector quantization and to increase the classification performance, a fuzzy approach based similarity was applied to implement the DHMM. The performance of the proposed Fuzzy DHMM (FDHMM) was compared with some methods such as DHMM, artificial neural network (ANN), neuro-fuzzy approaches and obtained better classification performance than these methods.