Linear trajectory segmental HMMs

The aim of the research described is to overcome important speech-modeling limitations of conventional hidden Markov models (HMMs), by developing a dynamic segmental HMM which models the changing pattern of speech over the duration of some phoneme-type unit. As a first step towards this goal, a static segmental HMM has been implemented and tested. This model reduces the influence of the independence assumption by using two processes to model variability due to long-term factors separately from local variability that occurs within a segment. Experiments have demonstrated that the performance of segmental HMMs relative to conventional HMMs is dependent on the quality of the system in which they are embedded. On a connected-digit recognition task for example, static segmental HMMs outperformed conventional HMMs for triphone systems but not for a vocabulary-independent monophone system. It is concluded that static segmental HMMs improve performance, as long as the system is such that the independence assumption is a major limiting factor.

Growth in Digital signal processing technology gives positives influences in human life. One of the branch of science that gives significant influence is digital speech processing. It can be expand into some applications that make human life easier. Digital speech processing is appropriate to speech recognition. Speech recognition is used to arrange the movement of remote control car. The remote control car will move according to our speech. This final project is closely related with speech recognition. The LPC (Linear Predictive Coding) method will extract the speech signal features and HMM (Hidden Markov Model) to modelling the speech signal are used. It is done by comparing model from extracting feature that is available in HMM modelling. Models will be used in the speech recognition process, if the models have highest level of conformity. The experiment has been done in two conditions, i.e ideal condition in room wih low noise level and unideal condition in room with noise. Result of the experiment from the whole sistem performace at ideal condition is 97,71% for people that have been inputed in the database, and 95,42% for people which have not been inputed in the database. Result of the experiment of database say it regulary is 97,14%. Result of the experiment of database at unideal condition is 54,28% for inputing word at noisy area, result of the experiment of database for inputing word at high frequency from siren voice is 98,57% and result of the experiment of inputing similar word with database is 97,6% Keywords: speech recognition, LPC, HMM

In this letter, we propose a new trajectory model for characterizing segmental features and their interaction based upon a general framework of hidden Markov models. Each segment, a sequence of frame vectors, is represented by a trajectory of observed vector sequences. This trajectory replaces the frame features in the segment and becomes the input of the segmental hidden Markov models (HMM's). In our approach, we adopt polynomial trajectory modeling to represent the trajectories using a new design matrix that includes transitional information on neighborhood acoustic events. To apply this trajectory to the segmental HMM, extra- and intrasegmental variations are modified to contain trajectory information. The presented model is regarded as an extension and generalization of conventional HMM, trajectory-based segmental HMM, and parametric trajectory models. The experimental results are reported on the TIMIT corpus and performance is shown to improve significantly over that of the conventional HMM.

This paper describes about speech recognition implemented on ATmega162 microcontroller. The word (voice command) in a speech signal is applied for controlling movement of a mobile robot. The mobile robot will move according to the voice command in the speech signal. There are five commands, which are Indonesian language, used to control the movement of mobile robot. They are “maju”, “mundur”, “kiri”, “kanan”, and “stop” used to command the mobile robot to move forward, move backward, turn left, turn right and stop respectively.The methods used in this research are Linear Predictive Coding (LPC) and Hidden Markov Model (HMM). LPC is used to extract word data from a speech signal. HMM is used to recognize the word pattern data, which are extracted from a speech signal. Sampling rate of the speech signal is 8 kHz and the speech signal is sampled for 0.5 seconds. Experiments were done in several variations of observation symbol number and number of sample. The experimental results show that the highest recognition rate that can be achieved by the system is 87%. The mobile robot can move in accordance with the voice command that is given to the robot.

This study analyzes the effect of stress in human and automatic stressed speech processing tasks for speech collected from non-professional speakers. The database of 33 keywords is collected under five stress conditions, namely, neutral, angry, happy, sad and Lombard from fifteen speakers. The first study is to understand the ability to identify stress by human and automatic speech processing. The average performance of human stress classification is 59.44%. The average performance of automatic stress classifier using Vector Quantization (VQ) and Hidden Markov model (HMM) is 54.65% and 56.02%, respectively. The second study has been done to understand the effect of stress in human and automatic speech recognition. The average performance of human stressed speech recognition is 99.60%. The automatic stressed speech recognition performance using VQ and HMM is 82.42% and 76.79%, respectively. Even in the non-professional speakers scenario, human performance is better than automatic processing. Also, automatic processing seem to show considerable degradation in performance that warrant development of new methods to handle stress information.

A hidden Markov model based classifier is proposed in this paper to perform automatic speech recognition using myoelectric signals from the muscles of vocal articulation. The classifier's resilience to temporal variance is compared to a linear discriminant analysis classifier that was used in a pervious study. Speech recognition was performed, using five channels of myoelectric signals, on isolated words from a 10-word vocabulary. Temporal variance was induced by temporally misaligning data from the test set, with respect to the training set. When compared to the LDA classifier, the hidden Markov model classifier demonstrated a markedly lower variation in classification error due to the temporal misalignment. Characteristics of the hidden Markov model MES classifier suggest that it would effectively complement a conventional acoustic speech recognizer, in a multi-modal speech recognition system.

A model-based framework of classification error rate estimation is proposed for speech and speaker recognition. It aims at predicting the run-time performance of a hidden Markov model (HMM) based recognition system for a given task vocabulary and grammar without the need of running recognition experiments using a separate set of testing samples. This is highly desirable both in theory and in practice. However, the error rate expression in HMM-based speech recognition systems has no closed form solution due to the complexity of the multi-class comparison process and the need for dynamic time warping to handle various speech patterns. To alleviate the difficulty, we propose a one-dimensional model-based misclassification measure to evaluate the distance between a particular model of interest and a combination of many of its competing models. The error rate for a class characterized by the HMM is then the value of a smoothed zero-one error function given the misclassification measure. The overall error rate of the task vocabulary could then be computed as a function of all the available class error rates. The key here is to evaluate the misclassification measure in terms of the parameters of environmental-matched models without running recognition experiments, where the models are adapted by very limited data that could be just the testing utterance itself. In this paper, we show how the misclassification measure could be approximated by first computing the distance between two mixture Gaussian densities, then between two HMMs with mixture Gaussian state observation densities and finally between two sequences of HMMs. The misclassification measure is then converted into classification error rate. When comparing the error rate obtained in running actual experiments and that of the new framework, the proposed algorithm accurately estimates the classification error rate for many types of speech and speaker recognition problems. Based on the same framework, it is also demonstrated that the error rate of a recognition system in a noisy environment could also be predicted.

Over the past few decades, designers are considering a range of applications ranging from mobile communications to automatic machine learning. Speeches are less commonly used in the electronic and computer field due to the complexity and variety of signals and sounds. By the use of modern algorithms and methods, speech signals are processed to recognize text. In this project, we will build a online speech to text engine. The program receives speech during the run through the microphone and uses sample speech to recognize the text. The known text can be saved to a file. This is being developed on Java platform using the eclipse workbench. Our speech-to-text program directly gets and converts speech into text. It can add other great applications, giving users a different choice of data input. A text-to-speech system can also improve accessibility of the system by providing data access options for users who are blind, deaf or disabled. Voice SMS application allows the user to record and convert spoken messages into an text message. The user can send messages to the phone number which is entered. Speech recognition is done through the Internet, connecting to Google's server. The application is based on input messages in English. Speech recognition uses a technique based on hidden Markov models (HMM - Hidden Markov Model). It is currently the most effective and flexible method of speech recognition.

In this dissertation, a novel approach of integrating template matching with statistical modeling is proposed to improve continuous speech recognition. Hidden Markov Modeling (HMMs) has been the dominant approach in statistical speech recognition since it provides a principled way of jointly modeling speech spectral variations and time dynamics. However, HMMs have the shortcoming of assuming the observations being independent within each state, which makes it ineffective in modeling the details of speech temporal evolutions that are important for characterizing nonstationary speech sounds. Template-based methods make comparisons between a test pattern and the templates derived from training data, and therefore they are able to capture speech dynamics and time correlation of speech frames better than HMM based methods. However, template matching requires large memory space and computational time since feature vectors of training data need to be stored in computer memory for access at the recognition stage, which is difficult in large vocabulary continuous speech recognition (LVCSR). Our proposed approach takes advantages of both statistical modeling and template matching, which overcomes the weakness of conventional template-based method and is feasible for LVCSR. We use multiple Gaussian Mixture Model (GMM) indices to represent each frame of speech templates, and define the template unit to be context-dependent phone segments (triphone context). We also use phonetic decision trees borrowed from those commonly used in HMMs to tie triphone templates and predict triphones unseen in training data. Two local distances, log likelihood ratio (LLR) and Kullback-Leibler (KL) divergence, are proposed for dynamic time warping (DTW) based template matching. In order to reduce computational complexity and storage space, we propose methods of minimum distance template selection (MDTS) and maximum log-likelihood template selection (MLTS), and investigate a template compression method on top of template selection to further improve recognition performance. The template based methods were used to rescore lattices generated by baseline HMMs on the tasks of TIMIT continuous phone recognition and teleheath LVCSR and experimental results demonstrated that the proposed approach of integrating template matching with statistical modeling significantly improved recognition performances over the HMM baselines. The template selection methods also provided significant recognition accuracy improvements over the HMM baseline while largely reducing the computation and storage complexities. When all templates or MDTS were used, using the LLR local distance obtained better recognition performance than the KL divergence local distance. For MLTS and template compression, KL divergence local distance provided better performance than the LLR local distance, and the template compression method made further improvements over KL based MLTS. Since the templates were constructed based on the GMM indices extracted from HMM baselines, we also validated the effectiveness of the proposed template methods based on enhanced HMM baselines. Experimental results showed that LLR based all template method was able to consistently improve TIMIT phone recognition accuracies based on four enhanced HMM baselines. Prosodic features such as duration, energy, and pitch can reflect longer span information of speech than conventional single frame vectors but they have commonly been ignored by HMMs. Template based methods provide possibilities to conveniently integrate prosodic features into speech recognition, which has not been well studied in the past. In this dissertation, we investigate combining template based methods with the speech prosodic features of duration, energy and pitch to further improve speech recognition accuracy. The scores of prosodic information were computed by a GMM based method and a non-parametric method, and the prosodic scores were combined with the acoustic scores in triphone template matching. Experimental results obtained on the telehealth task showed that prosodic information had positive effects on vowel sound recognition.

A low-power and real-time hidden Markov model (HMM) accelerator is proposed for gesture user interfaces on wearable smart devices. HMM algorithm is widely used for sequence recognitions such as speech recognition and gesture recognition due to its best-in-class recognition accuracy. However, the HMM algorithm incorporates high computational complexity and requires massive memory bandwidth for sequence matches. There have been studies on hardware acceleration of the HMM algorithm to resolve these issues, but they were focused on the speech recognition and did not incorporate the motion orientation capability required for the gesture recognition case. In this paper, we propose an HMM accelerator incorporating the motion orientation block for gesture recognitions on wearable devices. Binary search is exploited in the motion orientation to avoid the division and arctangent associated with the orientation and reduce its arithmetic complexity. In addition, gesture models are clustered in the gesture database to save the memory bandwidth by reducing memory transactions. Moreover, logarithmic arithmetic is used in Viterbi decoding in the HMM for more reduction in its complexity. Thanks to these schemes, this work achieves 25.6% power reduction compared with a plain hardware implementation of the gesture recognizing HMM.

In this paper, a brief overview derived out of detailed survey of speech recognition works reported in Indian languages is described. Robustness of speech recognition systems toward language variation is the recent trend of research in speech recognition technology. To develop a system which can communicate with human in any language like any other human is the foremost requirement in order to design appropriate speech recognition technology for one to all. India is a country which has vast linguistic variations among its billion plus population. Therefore, it provides a sound area of research toward language-specific speech recognition technology. From the beginning of the commercial availability of the speech recognition system, the technology has been dominated by the hidden Markov model (HMM) methodology due to its capability of modeling temporal structures of speech and encoding them as a sequence of spectral vectors. Most of the work done in Indian languages also uses HMM technology. However, from the last 10–15 years after the acceptance of neurocomputing as an alternative to HMM, artificial neural network (ANN)-based methodologies have started to receive attention for application in speech recognition. This is a trend worldwide as part of which few works have also been reported by a few researchers.

A speech recognition method which integrates a VQ (vector quantization)-distortion measure and a discrete HMM (hidden Markov model) is proposed. This VQ-distortion-based HMM uses a VQ-distortion measure at each state instead of the discrete output probability used by a discrete HMM. Although this method is regarded as a refined version of the VQ-distribution based recognition method proposed by D.K. Burton et al (IEEE Trans. vol. ASSP-33, no.4, p.837-49 of 1985), it is also considered as a special case of a mixtured distribution density HMM. The authors describe the relationship between the VQ-distortion-based HMM and conventional HMMs, and compare their speech recognition performance through experiments on speaker-dependent digit recognition. A recognition accuracy of 100% using the new method was obtained. >

Visual speech recognition is able to supplement the information of speech sound to improve the accuracy of speech recognition. A viseme, which describes the facial and oral movements that occur alongside the voicing of a particular phoneme, is a supposed basic unit of speech in the visual domain. As in phonemes, there are variations for the same viseme expressed by different persons or even by the same person. A classifier must be robust to this kind of variation. In this chapter, the author’s describe the Adaptively Boosted (AdaBoost) Hidden Markov Model (HMM) technique (Foo, 2004; Foo, 2003; Dong, 2002). By applying the AdaBoost technique to HMM modeling, a multi-HMM classifier that improves the robustness of HMM is obtained. The method is applied to identify context-independent and contextdependent visual speech units. Experimental results indicate that higher recognition accuracy can be attained using the AdaBoost HMM than that using conventional HMM.

Proposes a global acoustic segment model for characterizing fundamental speech sound units and their interactions based upon a general framework of hidden Markov models (HMM). Each segment model represents a class of acoustically similar sounds. The intra-segment variability of each sound class is modeled by an HMM, and the sound-to-sound transition rules are characterized by a probabilistic intersegment transition matrix. An acoustically-derived lexicon is used to construct word models based upon subword segment models. The proposed segment model was tested on a speaker-trained, isolated word, speech recognition task with a vocabulary of 1109 basic English words. In the current study, only 128 segment models were used, and recognition was performed by optimally aligning the test utterance with all acoustic lexicon entries using a maximum likelihood Viterbi decoding algorithm. Based upon a database of three male speakers, the average word recognition accuracy for the top candidate was 85% and increased to 96% and 98% for the top 3 and top 5 candidates, respectively. >

The movements of talkers' face, nose, mouth and throat are known to convey visual cues and represent several different kinds of informationl, and that can improve speech recognition rate, especially for persons with speech-impairments. We proposed a new speech recognition method using these visual features and hidden Markov model (HMM). Based on global optimisation, a new genetic algorithm (GA) for training HMM was proposed. Six chinese vowels were taken as the experimental data, ten handicapped speakers were taken as the testee. Recognition experiments show that the method is effective and high speed and accuracy for speech recognition. At present, the average recognition rate is 91.47% using improved HMM and 88.96% using the classic training HMM algorithm, So the features has very good robustness and the improved HMM is very good.