Continuous Hidden Markov Model Research Articles

Gaussian Model Optimization using Configuration Thread Control In CHMM Vocabulary Recognition

Abstract In vocabulary recognition using HMM(Hidden Markov Model) by model for the observation of a discrete probability distribution indicates the advantages of low computational complexity, but relatively low recognition rate has the disadvantage that require sophisticated smoothing process. Gaussian mixtures in order to improve them with a continuous probability density CHMM (Continuous Hidden Markov Model) model is proposed for the optimization of the library system. In this paper is system configuration thread control in recognition Gaussian mixtures model provides a model to optimize of the CHMM vocabulary recognition. The result of applying the proposed system, the recognition rate of 98.1% in vocabulary recognition, respectively. Key Words : Gaussian model, Model optimization, CHMM, Vocabulary Recognition, Library system ※본 논문은 2012년 가천대학교의 교내연구비 지원에 의한 결과임. (GCU-2012-R147) * 광운대학교 컴퓨터공학과 박사과정 ** 가천대학교 IT대학 인터랙티브미디어학과 교수(교신저자)논문접수: 2012년 7월 19일, 1차 수정을 거쳐, 심사완료: 2012년 8월 20일

A VLSI Architecture with Multiple Fast Store-Based Block Parallel Processing for Output Probability and Likelihood Score Computations in HMM-Based Isolated Word Recognition

This paper presents a memory-efficient VLSI architecture for output probability computations (OPCs) of continuous hidden Markov models (HMMs) and likelihood score computations (LSCs). These computations are the most time consuming part of HMM-based isolated word recognition systems. We demonstrate multiple fast store-based block parallel processing (MultipleFastStoreBPP) for OPCs and LSCs and present a VLSI architecture that supports it. Compared with conventional fast store-based block parallel processing (FastStoreBPP) and stream-based block parallel processing (StreamBPP) architectures, the proposed architecture requires fewer registers and less processing time. The processing elements (PEs) used in the FastStoreBPP and StreamBPP architectures are identical to those used in the MultipleFastStoreBPP architecture. From a VLSI architectural viewpoint, a comparison shows that the proposed architecture is an improvement over the others, through efficient use of PEs and registers for storing input feature vectors.

Traffic event classification at intersections based on the severity of abnormality

This paper proposes a novel traffic event classification approach using event severities at intersections. The proposed system basically learns normal and common traffic flow by clustering vehicle trajectories. Common vehicle routes are generated by implementing trajectory clustering with Continuous Hidden Markov Model. Vehicle abnormality is detected by observing maximum likelihoods of partial vehicle locations and velocities on underlying common route models. The second part of the work is based on extracting the severities of abnormality by deviation measurement using Coefficient of Variances method. By using abnormal event samples, two severity classes are built in order to recognize event severities by Support Vector Machines and k-Nearest Neighborhood algorithms. Experimental results show that the proposed model has high precision with satisfactory incident detection and event severity classification performance.

Trend forecasting of financial time series using PIPs detection and continuous HMM

Many machine learning methods in Artificial intelligence literature, such as Neural Networks, Genetic Algorithms, SVM, and Case-Based Reasoning, have been applied to forecast the financial market with high irregularity and uncertatinty. Among them, SVM is the most representative method that models and forecasts financial time series. However, SVM cannot reflect on the dynamic characteristic of financial time series effectively due to its superior generalization performance. And we cannot guarantee that the parameters of SVM have the optimum values, since they are locally searched owing to the limited time bound. These vulnerabilities are the main factors that degenerates the forecasting performance of SVM. On the other hand, continuous HMM can effectively capture the irregular and dynamic movement of financial time series with a non-stationary property, since it models a financial time series stochastically, rather than deterministically. Therefore, this paper suggests a new method that constructs the trend forecasting model of financial time series. It firstly detects PIPs indicating the significant turnabout of trend in each financial time series. And then the detected PIPs are used to construtct its trend forecasting model based on continuous HMM. In the experiment with various financial time series datasets we demonstrate its superiority.

CHMM for tool condition monitoring and remaining useful life prediction

To solve the problems of tool condition monitoring and prediction of remaining useful life, a method based on the Continuous Hidden Markov Model (CHMM) is presented. With milling as the research object, cutting force is taken as the monitoring signal, analyzed by wavelet packet theory to reduce noise and extract the energy feature of the signal as a basis for diagnosis. Then, CHMM is used to diagnose tool wear state. Finally, a Gaussian regression model is proposed to predict the milling tool’s remaining useful life after the test sample data are verified to be consistent with the Gaussian distribution based on a reliable identification of the milling tool wear state. The probability models of tool remaining useful life prediction could be established for tools with different initial states. For example, when an unknown state of milling force signal is delivered to the milling tool online diagnostic system, the state and the existing time of this state could be predicted by the established prediction model, and then, the average remaining useful life from the present state to the tool failure state could be obtained by analyzing the transfer time between each state in the CHMM. Compared to the traditional probabilistic model, which requires a large amount of test samples, the experimental cost is effectively reduced by applying the proposed method. The results from the experiment indicate that CHMM for tool condition monitoring has high sensitivity, requires less training samples and time, and produces results quickly. The method using the Gaussian process to accurately predict remaining life has ample potential for application to real situations.

An Implementation of Multimodal User Interface using Speech, Image and EOG

There have been many recent studies on gaze recognition system in the field of HCI (Human Computer Interaction). This system will be the most natural and intuitive HCI system due to the application of gaze direction or biomedical Signals. We propose a multimodal user interface system using the nine directional gaze recognition based on image, EOG (Electrooculography) signal and speech recognition. In this paper, we use DFA (Deterministic Finite Accepter), Haar-like feature and Adaboost algorithm, SVM (Support Vector Machine) for gaze recognition based on the EOG signal and image. Furthermore, the CHMM (Continuous Hidden Markov Model) based speech recognition system has been linked for inputting speech commands. The proposed multimodal user interface system solves the problem of constraint of single modal recognition systems. As a result, the proposed system achieves higher recognition performance and more natural interface using speech commands.

Driver Intention Recognition Method Using Continuous Hidden Markov Model

In order to make Intelligent Transportation System (ITS) work effectively, a driver intention recognition method is proposed. In this research, three different recognition models were developed based on Continuous Hidden Markov Model (CHMM), and could distinguish left and right lane change intention from normal lane keeping intention. Subjects performed lane change maneuvers and lane keeping maneuvers with driving simulator which simulated highway scenes, parameters that highly correlated with lane change behavior were collected and analyzed. A series of testings and comparisons were done to obtain the optimal model structure and feature set. Results show that, taking the steering wheel angel, steering wheel angle velocity and lateral acceleration as the optimal observation signals, the accuracy can achieve up 95%, and it proved very effective in terms of early intention recognition.

Design of a Low-Power Coprocessor for Mid-Size Vocabulary Speech Recognition Systems

Speech recognition systems have gained popularity in consumer electronics. This paper presents a custom-designed coprocessor for output probability calculation (OPC), which is the most computation-intensive processing step in continuous hidden Markov model (CHMM)-based speech recognition algorithms. To save hardware resource and reduce power consumption, a polynomial addition-based method is used to compute add-log instead of the traditional look-up table-based method. In addition, the optimal tradeoff between speech processing delay, energy consumption, and hardware resources is explored for the coprocessor. The proposed coprocessor has been implemented and tested in Xilinx Spartan-3A DSP XC3SD3400A, and also validated using the standard-cell-based approach in IBM 0.13 μm technology. To implement an entire speech recognition system, SAMSUNG S3C44b0X (containing an ARM7) is used as the micro-controller to execute the rest of speech processing. Tested with a 358-state 3-mixture 27-feature 800-word HMM, S3C44b0X operates at 40 MHz and coprocessor at 10 MHz to meet the real-time requirement, and the recognition accuracy is 95.2%. Power consumption of the micro-controller is 10 mW, and that of the coprocessor 15.2 mW. The overall speech recognition system achieves the lowest energy consumption per word recognition among many reported designs. Experiment and analysis show that the speech recognition system based on the proposed coprocessor is especially suitable for mid-size vocabulary (100-1000 words) recognition tasks.

Glottal Excitation Feature based Gender Identification System using Ergodic HMM

In this paper, through different experimental studies it is demonstrated that the time varying glottal excitation component of speech can be exploited for text independent gender recognition studies. Linear prediction (LP) residual is used as a representation of excitation information in speech. The gender-specific information in the excitation of voiced speech is captured using the Hidden Markov Models (HMMs). The decrease in the error during training and recognizing genders during testing phase close to 100 % accuracy demonstrates that the excitation component of speech contains gender-specific information and is indeed being effectively captured by continuous Ergodic HMM. A gender recognition study using gender specific features for different HMM states, mixture components, size of testing data on the performance of the gender recognition is evaluated. We demonstrate the gender recognition studies on TIMIT database. KeywordsGender, Hidden Markov Model (HMM); LPC; MFCC

A high-performance text-independent speaker identification of Arabic speakers using a CHMM-based approach

This paper reports an approach that depends on Continuous Hidden Markov Models (CHMMs) to identify Arabic speakers automatically from their voices. The Mel-Frequency Cepstral Coefficients (MFCCs) were selected to describe the speech signal. The general Gaussian density distribution HMM is developed for the CHMM system. Ten Arabic speakers were used to evaluate our proposed CHMM-based engine. The identification rate was found to be 100% during text dependent experiments. However, for the text-independent experiments, the identification rate was found to be 80%.

결함 데이터를 필요로 하지 않는 연속 은닉 마르코프 모델을 이용한 새로운 기계상태 진단 기법

Model based machine condition diagnosis methods are generally using a normal and many failure models which need sufficient data to train the models. However, data, especially for failure modes of interest, is very hard to get in real applications. So their industrial applications are either severely limited or impossible when the failure models cannot be trained. In this paper, continuous hidden Markov model(CHMM) with only a normal model has been suggested as a very promising machine condition diagnosis method which can be easily used for industrial applications. Generally hidden Markov model also uses many pattern models to recognize specific patterns and the recognition results of CHMM show the likelihood trend of models. By observing this likelihood trend of a normal model, it is possible to detect failures. This method has been successively applied to arc weld defect diagnosis. The result shows CHMM's big potential as a machine condition monitoring method.

Driver Intention Recognition Method Using Continuous Hidden Markov Model

Driver Intention Recognition Method Using Continuous Hidden Markov Model

動作パターンとシンボルを相互変換する原始シンボル空間における動作パターンの内挿・外挿

Simple and easy human-interfaces for designing motion patterns of humanoids or human-like computer graphics have been required. One of the promising approaches is associating motion patterns with some symbol representations, and then generating new motion patterns by synthesizing those symbol representations. The proto-symbol space method based on continuous hidden Markov model (HMM) was such an approach. This method not only abstracts a symbol representation called proto-symbol but also generates new motion patterns by interpolating two proto-symbols in the proto-symbol space constructed with Kullback-Leibler divergence of the proto-symbols. Extrapolation of the proto-symbols as well as synthesis of HMMs that have different number of nodes, however, could not be achieved. In this paper, we propose an algorithm to solve these issues. We further discuss identification of synthesized proto-symbols and estimation of synthetic coefficients. To acquire accurate estimations, Bhattacharyya distance is introduced in the proto-symbol space construction. Finally, we show the feasibility of our approach with simulation experiments by a humanoid robot. This proto-symbol space could be a first synthetic step toward understanding symbolization mechanism according to Deacon's symbol developmental model.

Detection of unusual optical flow patterns by multilevel hidden Markov models

The analysis of motion information is one of the main tools for the understanding of complex behaviors in video. However, due to the quality of the optical flow of low-cost surveillance camera systems and the complexity of motion, new robust image-processing methods are required to generate reliable higher-level information. In our novel approach there is no need for tracking objects (vehicles, pedestrians) in order to recognize anomalous motion, but dense optical flow information is used to construct mixtures of Gaussians, which are analyzed temporally. We create a multilevel model, where low-level states of non-overlapping image regions are modeled by continuous hidden Markov models (HMMs). From low-level HMMs we compose high-level HMMs to analyze the occurrence of the low-level states. The processing of large numbers of data in traditional HMMs can result in a precision problem due to the multiplication of low probability values. Thus, besides introducing new motion models, we incorporate a scaling technique into the mathematical model of HMMs to avoid precision problems and to get an effective tool for the analysis of large numbers of motion vectors. We illustrate the use of our models with real-life traffic videos.

Development of a silent speech interface driven by ultrasound and optical images of the tongue and lips

This article presents a segmental vocoder driven by ultrasound and optical images (standard CCD camera) of the tongue and lips for a “silent speech interface” application, usable either by a laryngectomized patient or for silent communication. The system is built around an audio–visual dictionary which associates visual to acoustic observations for each phonetic class. Visual features are extracted from ultrasound images of the tongue and from video images of the lips using a PCA-based image coding technique. Visual observations of each phonetic class are modeled by continuous HMMs. The system then combines a phone recognition stage with corpus-based synthesis. In the recognition stage, the visual HMMs are used to identify phonetic targets in a sequence of visual features. In the synthesis stage, these phonetic targets constrain the dictionary search for the sequence of diphones that maximizes similarity to the input test data in the visual space, subject to a concatenation cost in the acoustic domain. A prosody-template is extracted from the training corpus, and the final speech waveform is generated using “Harmonic plus Noise Model” concatenative synthesis techniques. Experimental results are based on an audiovisual database containing 1 h of continuous speech from each of two speakers.

Efficient codebooks for fast and accurate low resource ASR systems

Today, speech interfaces have become widely employed in mobile devices, thus recognition speed and resource consumption are becoming new metrics of Automatic Speech Recognition (ASR) performance. For ASR systems using continuous Hidden Markov Models (HMMs), the computation of the state likelihood is one of the most time consuming parts. In this paper, we propose novel multi-level Gaussian selection techniques to reduce the cost of state likelihood computation. These methods are based on original and efficient codebooks. The proposed algorithms are evaluated within the framework of a large vocabulary continuous speech recognition task.

Handwritten word-spotting using hidden Markov models and universal vocabularies

Handwritten word-spotting is traditionally viewed as an image matching task between one or multiple query word-images and a set of candidate word-images in a database. This is a typical instance of the query-by-example paradigm. In this article, we introduce a statistical framework for the word-spotting problem which employs hidden Markov models (HMMs) to model keywords and a Gaussian mixture model (GMM) for score normalization. We explore the use of two types of HMMs for the word modeling part: continuous HMMs (C-HMMs) and semi-continuous HMMs (SC-HMMs), i.e. HMMs with a shared set of Gaussians. We show on a challenging multi-writer corpus that the proposed statistical framework is always superior to a traditional matching system which uses dynamic time warping (DTW) for word-image distance computation. A very important finding is that the SC-HMM is superior when labeled training data is scarce—as low as one sample per keyword—thanks to the prior information which can be incorporated in the shared set of Gaussians.

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.

Real-Time Continuous Phoneme Recognition System Using Class-Dependent Tied-Mixture HMM With HBT Structure for Speech-Driven Lip-Sync

This work describes a real-time lip-sync method using which an avatar's lip shape is synchronized with the corresponding speech signal. Phoneme recognition is generally regarded as an important task in the operation of a real-time lip-sync system. In this work, the use of the Head-Body-Tail (HBT) model is proposed for the purpose of more efficiently recognizing phonemes which are variously uttered due to co-articulation effects. The HBT model effectively deals with the transition parts of context-dependent models for small-sized vocabulary tasks. These models provide better recognition performance than general context-dependent or context-independent models for the task of digit or vowel recognition. Moreover, each phoneme is categorized into one among four classes and the class-dependent codebook is generated to further improve the performance. Additionally, for the clear representation of the context dependency information in the transient parts, some Gaussians are excluded from class-dependent codebook. The proposed method leads to a lip-sync system that performs at a level that is similar to previous designs based on HBT and continuous hidden Markov models (CHMMs). However, our method reduces the number of model parameters by one-third and enables real-time operation.

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%).