Hidden Markov Model Algorithm Research Articles

Online learning system for English speech automatic recognition based on hidden Markov model algorithm and conditional random field algorithm

E-learning provides a flexible way of learning, which can provide corresponding learning content and tasks according to students’ interests, abilities and learning progress, so that students can choose learning content and learning progress more independently. In traditional teaching methods, teachers need to spend a lot of time and energy to evaluate students’ learning. The e-learning system can automatically record students’ learning behavior and learning results, and give feedback and evaluation in time during the learning process. Automatic speech recognition systems have been widely applied in various fields, and the demand for more user-friendly and convenient user operating systems is becoming increasingly urgent. This article proposes an English automatic speech recognition system based on hidden Markov model algorithm and conditional random field algorithm to improve recognition accuracy and stability, and applies it to the computer field. Label inference and optimization for speech recognition were carried out using the conditional random field algorithm. By establishing the correlation between features and labels, label inference was performed on the speech signal, which was compared with existing speech recognition systems to evaluate whether the recognition accuracy and stability of the new system had been improved. The results show that the English automatic speech recognition system based on hidden Markov model algorithm and conditional random field algorithm exhibits high performance in terms of recognition accuracy and stability. Compared with traditional speech recognition systems, this system has made significant improvements in the application of computer science.

On the definitions of hidden Markov models

At least three probabilistic definitions of hidden Markov models (HMMs) have been used frequently in the literature. Unfortunately, one of these definitions shows fatal flaws, however nowadays a lot of literature still uses this definition. The aim of this paper is on one hand to specifically point out one such fatal flaw (in terms of deriving the well-known forward-backward algorithm), and on the other hand to list key properties of HMMs under the other two plausible (and equivalent) probabilistic definitions for further developments. As applications, we rigorously layout forward-backward algorithms for inhomogeneous HMMs and hidden reciprocal models, and fully present connections between HMMs and undirected graphical models which are not mentioned anywhere in the literature, to the best of our knowledge.

Forecasting wildlife movement with spatialcapture–recapture

AbstractWildlife movement is an important process affecting species population biology and community interactions in myriad ways. Studies of wildlife movement have focused on retrospectively estimating movements of small numbers of individuals by outfitting them with GPS and telemetry tags. Recent developments in spatial capture–recapture modelling permit the integration of movement models that can estimate the movement of untagged and undetected individuals. Additionally, hidden Markov movement models provide a framework for forecasting individuals' movements, which may be valuable in the conservation of threatened species facing risks that vary across space and time.We describe maximum likelihood estimators for spatial capture–recapture models integrated with simple, biased and correlated random walk movement models formulated as hidden Markov models. Additionally, we demonstrate how to forecast wildlife movement based on these models and hidden Markov model algorithms. We conducted a simulation study to test the performance of the models' abundance estimators and movement forecasts when fit to data simulated under different movement models. We also fit the models to spatial capture–recapture data collected on North Atlantic right whales off the Atlantic Coast of the southeastern United States.Random walk movement models improved abundance estimation and movement forecasts in our simulation study and received greater support from the data in the right whale case study than did activity centre movement models.Forecasts of wildlife movement made under integrated spatial capture–recapture movement models will be most valuable when individuals have been observed recently, when sampling for individuals is extensive and efficient, and when the scale of individuals' movements is small relative to the scale of the study area and sampling process.

The application of HMM algorithm based music note feature recognition teaching in universities

With the development of information technology, computer technology is also gradually applied to the teaching activities of art education, and the use of multimedia technology to assist music teaching has become one of the hot research areas in universities. In order to better cultivate university students' musical exploration ability and creativity, a music note feature recognition teaching model based on Hidden Markov Model (HMM) algorithm is studied and optimized in universities by using genetic algorithm based on HMM algorithm. The music note feature recognition teaching model studied in this article combines computer multimedia technology, signal processing technology, and music theory, and uses computers to simulate the process of human cognition and analysis of music. And in this article, a music note recognition system was constructed using the features of sound level contours combined with the HMM algorithm. The data extracted during music recognition was compressed using the energy compression feature of sound level contours. At the same time, maximum likelihood estimation was used to find the optimal chord sequence, i.e., the optimal path, for the input signal. In the experimental results, the minimum value of the objective function was about 0.739 when the variance probability and crossover probability were 0.02 and 0.6, respectively. In the results, the HMM algorithm-based music note recognition model can improve the quality of music teaching and has some potential for application in the field of music teaching.

Prediction of Cyberspace Security Data Based on the Markov Chain Model

Abstract In the era of people’s access to information, the network has undergone tremendous development, the connection patterns of network attacks have become endless, and the number of attack network methods is increasing year by year. Based on the Markov model, the author studies the problem of network security authentication and aims to determine the truth of network security. In this way, network security managers can identify current network security issues, and we can gain a better understanding of the current network situation so that we can update our current security network promptly. The development process proposed by the author is simulated and the obtained results are all compared and evaluated. The experimental results show that the hidden Markov model algorithm is improved, the network security situation is better recognized, and the accuracy of the n network security problem is better known. At the same time, it is known to predict network security problems in the future according to the application model of predicting network security problems. Tests have shown that the development process is feasible and effective.

Compressed computations using wavelets for hidden Markov models with continuous observations.

Compression as an accelerant of computation is increasingly recognized as an important component in engineering fast real-world machine learning methods for big data; c.f., its impact on genome-scale approximate string matching. Previous work showed that compression can accelerate algorithms for Hidden Markov Models (HMM) with discrete observations, both for the classical frequentist HMM algorithms-Forward Filtering, Backward Smoothing and Viterbi-and Gibbs sampling for Bayesian HMM. For Bayesian HMM with continuous-valued observations, compression was shown to greatly accelerate computations for specific types of data. For instance, data from large-scale experiments interrogating structural genetic variation can be assumed to be piece-wise constant with noise, or, equivalently, data generated by HMM with dominant self-transition probabilities. Here we extend the compressive computation approach to the classical frequentist HMM algorithms on continuous-valued observations, providing the first compressive approach for this problem. In a large-scale simulation study, we demonstrate empirically that in many settings compressed HMM algorithms very clearly outperform the classical algorithms with no, or only an insignificant effect, on the computed probabilities and infered state paths of maximal likelihood. This provides an efficient approach to big data computations with HMM. An open-source implementation of the method is available from https://github.com/lucabello/wavelet-hmms.

A reaction network scheme for hidden Markov model parameter learning.

With a view towards artificial cells, molecular communication systems, molecular multiagent systems and federated learning, we propose a novel reaction network scheme (termed the Baum-Welch (BW) reaction network) that learns parameters for hidden Markov models (HMMs). All variables including inputs and outputs are encoded by separate species. Each reaction in the scheme changes only one molecule of one species to one molecule of another. The reverse change is also accessible but via a different set of enzymes, in a design reminiscent of futile cycles in biochemical pathways. We show that every positive fixed point of the BW algorithm for HMMs is a fixed point of the reaction network scheme, and vice versa. Furthermore, we prove that the 'expectation' step and the 'maximization' step of the reaction network separately converge exponentially fast and compute the same values as the E-step and the M-step of the BW algorithm. We simulate example sequences, and show that our reaction network learns the same parameters for the HMM as the BW algorithm, and that the log-likelihood increases continuously along the trajectory of the reaction network.

Detection of Human Protein Structures by Select Deep Learning Models and Dynamic Systems

Peptide bonds bind amino acids together to form proteins. Because of the way amino acids fold, proteins have a three-dimensional structure. Increased data, or samples, improves machine learning models that learn from examples. ANNs, a single-layer learning system, require sophisticated systems and time to learn massive volumes of data over time. Today, the multi-layer deep learning technique is preferred. Deep learning uses artificial neural networks. This part of the work provides technical details about the deep learning methods used. This research's deep learning methods include CNN, Hidden Markov Model (HMM), Recurrent Weighted Average (RWA), and Conditional Random Fields (CRF). The training and testing results of these models are presented in this work. For this work, the CNN, HMM, RWA, and CRF algorithms were all compared using the CB513 dataset. In addition, each method was analyzed and contrasted with other research conducted previously. In this particular research endeavour, the respective success rates of the four models that were built for predicting the protein secondary structure were as follows: 82.54%, 81.06%, 81.10%, and 81.48%. The models and working environment produced in this study can be used to predict protein secondary structure quickly. In deep learning experiments, data amount affects learning. Increasing data will test the study's models

Occupancy states forecasting with a hidden Markov model for incomplete data, exploiting daily periodicity

To automatize HVAC energy savings in buildings, it is useful to forecast the occupants’ behaviour. This article deals with such a forecasting problem by exploiting the daily periodicity of the input variables and the ability of the proposed model to learn from missing data. We propose a case study of occupancy behaviour, for which only a history of indirect information is available: CO2 and decibel levels. The unobserved classes are modeled by a Hidden Markov Model (HMM) which emits sensors’ data and periodic variables. The fact that Baum-Welch’s algorithm for HMMs is an instance of the EM algorithm allows the model to learn from incomplete data. In this case, the missing data are the class labels at every instant (past and future) and the sensors’ data in the future, but the time periodic variables are known at every instant: this enables the model to classify both the past instants (history) and the future instants (forecasting). Our method is tested on simulated data sets with 5%, 15%, 25% and 50% missing data that provide respectively an average classification accuracy of 93%, 89%, 85% and 80%. Then, experiments on one-day and one-week forecasting scenarios achieved an average accuracy of 82% and 81%. The proposed method is also applied on a real-life data set which seems to provide promising results.

A Spectral Method of Moments for Hierarchical Imitation Learning

Recent empirical success has led to a rise in popularity of the options framework for Hierarchical Reinforcement Learning (HRL). This framework tackles the scalability problem in Reinforcement Learning (RL) by introducing a layer of abstraction (i.e. high-level options) over the (low-level) decision process. Hierarchical Imitation Learning (HIL) is the problem of learning low-level and high-level policies within HRL from expert demonstrations consisting only of the low-level actions and states, with the high-level options being hidden (or latent). Due to the latent options, recent work on HIL has focused on the development of Expectation-Maximization (EM) algorithms inspired by approaches such as the celebrated Baum-Welch algorithm for hidden Markov models (HMMs). In this work, we take a different approach and derive a new HIL framework inspired by the spectral method of moments for HMMs. The method of moments offers global and consistent convergence under mild regulatory conditions, whilst only requiring one sweep through the data set of state and action pairs, giving it a competitive run time.

DEEP REINFORCEMENT LEARNING WITH HIDDEN MARKOV MODEL FOR SPEECH RECOGNITION

Nowadays, many applications uses speech recognition especially the field of computer science and electronics, Speech Recognition (SR) is the interpretation of words spoken into a text. It is also known as Speech-To-Text (STT) or Automatic-Speech-Recognition(ASR), or just Word-Recognition(WR). The Hidden-Markov-Model (HMM) is a type of Markov model, which means that the future state of the model depends on the current state, not on the entire history of the system and the goal of HMM is to learn a sequence of hidden states from a set of known states. The Long-Short-Time-Memory (LSTM) network is a type of Recurrent Neural Network (RNN) that can learn long-term dependencies between time steps of sequence data. The LSTM network is trained by the network in order to predict the values of subsequent time steps in a series-to-series regression. Deep Neural Network (DNN) models are better classifiers than Gaussian Mixture Models (GMMs), they can generalize much better with a smaller number of parameters over complex distributions. They model distributions of different classes jointly, called “distributed” learning, or, more properly “tied” learning. This work is aimed at developing a speech recognition model that will predict isolated speech of some selected fruits in Hausa, Igbo and Yoruba language by using the predicting power of Mel-Frequency-Cepstral-Coefficient (MFCC), LSTM and HMM algorithms. The findings of the study would improve the development of better automatic speech applications systems and would benefit the academic and research community in the field of Natural Language Processing.

Validating continuous gravitational-wave candidates from a semicoherent search using Doppler modulation and an effective point spread function

Following up large numbers of candidates in continuous gravitational-wave searches presents a challenge, particularly in regard to computational power and the time required to manually scrutinize each of the candidates. It is important to design and test good follow-up procedures that are safe (i.e., minimize false dismissals) and computationally efficient across many search configurations. We investigate two follow-up procedures, or ``vetoes,'' both of which exploit the Doppler modulation predicted in astrophysical signals. In particular, we introduce the concept of using an effective point spread function as part of our veto criteria. We take advantage of a well-established semicoherent search algorithm based on a hidden Markov model to study various search configurations and to generalize the veto criteria by considering the overall veto performance in terms of efficiency and safety. The results can serve as a guideline for follow-up studies in future continuous gravitational-wave searches using a hidden Markov model algorithm. The results also apply qualitatively to other semicoherent search algorithms.

A hidden Markov optimization model for processing and recognition of English speech feature signals

Abstract Speech recognition plays an important role in human–computer interaction. The higher the accuracy and efficiency of speech recognition are, the larger the improvement of human–computer interaction performance. This article briefly introduced the hidden Markov model (HMM)-based English speech recognition algorithm and combined it with a back-propagation neural network (BPNN) to further improve the recognition accuracy and reduce the recognition time of English speech. Then, the BPNN-combined HMM algorithm was simulated and compared with the HMM algorithm and the BPNN algorithm. The results showed that increasing the number of test samples increased the word error rate and recognition time of the three speech recognition algorithms, among which the word error rate and recognition time of the BPNN-combined HMM algorithm were the lowest. In conclusion, the BPNN-combined HMM can effectively recognize English speeches, which provides a valid reference for intelligent recognition of English speeches by computers.

Optimal determination of hidden Markov model parameters for fuzzy time series forecasting

This paper presents Fuzzy Time Series (FTS) forecasting technique using Hidden Markov Model (HMM) optimized by Particle Swarm Optimization (PSO) and Genetic algorithm (GA). One of the major limitations of HMM is the lack of efficient method for HMM parameter estimation. Traditional methods like Baum Welch Algorithm (BWA) have been used to address the associated problem with HMM model. The BWA in itself does not truly capture the fuzziness in natural data resulting in the HMM algorithm into local minima. To address this challenge, this paper formulates the HMM parameter estimation problem into an optimization problem optimized by GA and PSO algorithms. To solve the problem of insufficiency in data allied with the HMM model, we adopted a method called smoothing to reduce the occurrence of zero in the observation events. Monte Carlo simulation is used at the end of the forecast to maintain stability and efficiency of the developed approach. The performance of developed model is evaluated using data of daily average temperature and cloud density of Taipei Taiwan, the Internet traffic data of Ahmadu Bello University (ABU) and Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX). Experimental results showed that the proposed forecasting models has a good forecasting accuracy when compared with existing models.

Hidden Markov Modeling with HMMTeacher.

Is it possible to learn and create a first Hidden Markov Model (HMM) without programming skills or understanding the algorithms in detail? In this concise tutorial, we present the HMM through the 2 general questions it was initially developed to answer and describe its elements. The HMM elements include variables, hidden and observed parameters, the vector of initial probabilities, and the transition and emission probability matrices. Then, we suggest a set of ordered steps, for modeling the variables and illustrate them with a simple exercise of modeling and predicting transmembrane segments in a protein sequence. Finally, we show how to interpret the results of the algorithms for this particular problem. To guide the process of information input and explicit solution of the basic HMM algorithms that answer the HMM questions posed, we developed an educational webserver called HMMTeacher. Additional solved HMM modeling exercises can be found in the user’s manual and answers to frequently asked questions. HMMTeacher is available at https://hmmteacher.mobilomics.org, mirrored at https://hmmteacher1.mobilomics.org. A repository with the code of the tool and the webpage is available at https://gitlab.com/kmilo.f/hmmteacher.

Coping with the liquidity crisis: a new dynamic quota readjustment scheme for carbon markets.

Because of insufficient liquidity, prices in the carbon market are more vulnerable to unexpected events, for which the impact duration lasts longer than that of the general market. The root reason for this phenomenon lies in the irrationality of quota distribution. The existing quota adjustment schemes and policies, e.g., the market stability reserve (MSR) and some recent adjustment measures, have poor timeliness and effectiveness, which has increased the risk of market crashes. Using the Hidden Markov Model (HMM), this paper develops a new dynamic quota adjustment scheme that can rapidly reduce the risk of quota supply by bridging quota price and quantity with price feedback as a response signal. To achieve this, we integrated the HMM algorithm and a two-step quota adjustment model by setting price thresholds and then connected the quota adjustment transition matrix and historical quota price. By comparing the MSR from 2013 to 2018, our scheme will help mitigate risks in quota price because the HMM can show the actual impact of price feedback on quota adjustment with merits of steady quota price and timely supply optimization. Moreover, our scheme, which recalculates the transition matrix, can be applied in other mature carbon markets.

ROHMM-A flexible hidden Markov model framework to detect runs of homozygosity from genotyping data.

Runs of long homozygous (ROH) stretches are considered to be the result of consanguinity and usually contain recessive deleterious disease-causing mutations. Several algorithms have been developed to detect ROHs. Here, we developed a simplealternative strategy by examining X chromosome non-pseudoautosomal region to detect the ROHs from next-generation sequencing data utilizing the genotype probabilities and the hidden Markov model algorithm as a tool, namely ROHMM. It is implemented purely in java and contains both a command line and a graphical user interface. We tested ROHMM on simulated data as well as real population data from the 1000G Project and a clinical sample. Our results have shown that ROHMM can perform robustly producing highly accurate homozygosity estimations under all conditions thereby meeting and even exceeding the performance of its natural competitors.

Learning hidden Markov models from aggregate observations

In this paper, we propose an algorithm for estimating the parameters of a time-homogeneous hidden Markov model (HMM) from aggregate observations. This problem arises when only the population level counts of the number of individuals at each time step are available, and one seeks to learn the individual HMM from these observations. Our algorithm is built upon the classical expectation–maximization algorithm and the recently proposed aggregate inference algorithm (Sinkhorn belief propagation). We present the parameter learning algorithm for two different settings of HMMs: one with discrete observations and one with continuous observations, and the algorithm exhibits convergence guarantees in both cases. Moreover, our learning framework naturally reduces to the standard Baum–Welch learning algorithm for HMMs when the population size is 1. The efficacy of our algorithm is demonstrated through several numerical experiments.

Application Research on Roller Skater Detection, Tracking, and Trajectory Prediction Based on Video Stream

With the continuous progress of science and technology, the sport of roller skating has developed rapidly and the technical level of the game has become higher and higher. Its sport performance has been rapidly improved. However, China’s roller skating is relatively late, and there is still a certain gap compared with many Western developed countries. In order to improve the performance of China’s roller skating, this study takes the representative Chinese and foreign excellent speed skaters as the research object and compares the sprinting technology of Chinese and foreign excellent speed skaters by using image measurement and image analysis to obtain the kinematic parameters and data of the athletes’ sprinting technology in the competition state. In view of the problem that the current video target tracking algorithm is easy to follow multiple targets, a video multiobject detection and tracking algorithm with improved tracking learning detection (TLD) is studied with the skater in the video as the research object. For the lost target, the prediction function of Kalman filter algorithm is used to track the trajectory of the typical target in the video, and the trajectory tracked by Kalman filter algorithm is used to compensate the lost part of TLD algorithm, so as to obtain the complete trajectory of the typical target in the video to improve the accuracy of video multiobject tracking. Since the existing trajectory prediction algorithms have the limitation of poor accuracy, a social-long short-term memory (Social-LSTM) network-based video typical target trajectory prediction algorithm is proposed to predict the trajectory sequences of typical targets to be detected by incorporating the contextual environment information and the interaction relationship between multiple target trajectories into the Social-LSTM network. The simulation results show that the proposed trajectory prediction algorithm outperforms the traditional LSTM algorithm, Hidden Markov Model Algorithm, and Hybrid Gaussian model algorithm, which is helpful to improve the accuracy of video roller skater target trajectory prediction, and the tracking success rate is 0.98.

A study on the algorithm of ultrasonic detection and recognition based on DAG‐SVMs mixed HMM of teleoperation gestures for intelligent manufacturing devices

Remote control for the position and status of a machine or an equipment can often be teleoperated by gestures in an intelligent manufacturing environment. In order to solve the problems that gestures with two directions such as left and right cannot be detected by single ultrasonic frequency, double different ultrasonic frequencies are used to detect gestures by the Doppler shift, and an algorithm of the recognition gesture based on the DAG-SVMs mixed Hidden Markov Model (HMM) is proposed to identify and classify the extracted feature sequences. Thus, four more types of gestures are expanded other than that of reading display screen information, and the comparative experiments to classify and recognise gestures of teleoperation are made with DAG-SVMs, the HMM, the DAG-SVMs mixed HMM, and other improved HMM algorithms. The test results have shown that the mean rate of gesture recognition for the algorithm based on the DAG-SVMs mixed HMM is 94.917%, which is 9.497% higher than that of the unimproved HMM, and its recognition accuracy of complex teleoperation gestures is improved by 2.3% compared with other improved HMM algorithms. The experimental results show that the DAG-SVMs mixed HMM algorithm has a good effect on recognition for the gestures of teleoperation and it can perform gesture recognition accurately.