In this paper, we discuss the problem of how human skill can be represented as parametric model using a hidden Markov model (HMM), and how a HMM-based skill model can be used to learn human skill. HMM is feasible to characterize two stochastic processes - measurable action and immeasurable mental states - which oze involved in the skill learning. We formul ted the learning problem as a multi-dimensional HMM and developed programming system which serve as a skill learning testbed for a variety of applications. Based on 'the most likely performance" criterion, we can select the best action sequence from a]l previously measured action data by modeling the skill as HMM. This selection process can be updated in real-time by feeding new action data and modifying HMM parameters. We address the imp]emcntatlon of the proposed method in a teleoperation-controlled space robot. An operator specifies the control command by a hand controller for the task of exchanging Orbit Replaceable Unit, and the robot learns the operation skill by selecting the sequence which represents the most likely performance of the operator. The skill is learned in Caxtesian space, joint space, and velocity domain. The experimental results demonstrate the feasibility of the proposed method in learning human skill and teleopertion control. The learning is significant in eliminating sluggish motion and correcting the motion command which the operator mistakenly generates.

In large-vocabulary speech recognition, the decoder often encounters triphones that are not covered in the training data. These unseen triphones are usually represented by corresponding diphones or context independent monophones. We propose to use decision-tree based senones to generate needed senonic baseforms for unseen triphones. A decision tree is built for each individual Markov state of each phone, and the leaves of the trees constitute the senone codebook. To find the senone a Markov state of any triphone is associated with, we traverse the corresponding tree until we reach a leaf node, where a senone is represented. We used the DARPA 5,000-word speaker-independent Wall Street Journal dictation task to evaluate the proposed method. The word error rate was reduced by 11% when unseen triphones were modeled by the decision-tree based senones. When there were at least 5 unseen triphones in each test utterance, the error rate could be reduced by more than 20%. This research was spons...

This thesis examines how artificial neural networks can benefit a large vocabulary, speaker independent, continuous speech recognition system. Currently, most speech recognition systems are based on hidden Markov models (HMMs), a statistical framework that supports both acoustic and temporal modeling. Despite their state-of-the-art performance, HMMs make a number of suboptimal modeling assumptions that limit their potential effectiveness. Neural networks avoid many of these assumptions, while they can also learn complex functions, generalize effectively, tolerate noise, and support parallelism. While neural networks can readily be applied to acoustic modeling, it is not yet clear how they can be used for temporal modeling. Therefore, we explore a class of systems called NN-HMM hybrids, in which neural networks perform acoustic modeling, and HMMs perform temporal modeling. We argue that a NN-HMM hybrid has several theoretical advantages over a pure HMM system, including better acoustic ...

Abstract—This paper presents a method for effectively using unlabeled sequential data in the learning of hidden Markov models (HMMs). With the conventional approach, class labels for unlabeled data are assigned deterministically by HMMs learned from labeled data. Such labeling often becomes unreliable when the number of labeled data is small. We propose an extended Baum-Welch (EBW) algorithm in which the labeling is undertaken probabilistically and iteratively so that the labeled and unlabeled data likelihoods are improved. Unlike the conventional approach, the EBW algorithm guarantees convergence to a local maximum of the likelihood. Experimental results on gesture data and speech data show that when labeled training data are scarce, by using unlabeled data, the EBW algorithm improves the classification performance of HMMs more robustly than the conventional naive labeling (NL) approach. Index Terms—Unlabeled data, sequential data, hidden Markov models, extended Baum-Welch algorithm. æ 1

Abstract not found

An approach is proposed for partial tying of states of tiedmixture hidden Markov models. To facilitate tying at the substate level, the state emission probabilities are constructed in two stages, or equivalently, are viewed as a "mixture of mixtures of Gaussians."This paradigm allows, and is complemented with, an optimization technique to seek the best complexity-accuracy tradeoff solution, which jointly exploits Gaussian density sharing and sub-state tying. Experimental results on the E-set show that the classification error rate is reduced by over 20% compared to standard Gaussian sharing and whole-state tying. The approach is then embedded within the recently developed procedure of combined parameter training and reduction technique. Experiments with the overall technique show that the error rate is further reduced by 8%.

The subject of this paper is the integration of the traditional vector quantizer (VQ) and discrete hidden Markov models (HMM) combination in the mixture emission density framework commonly used in automatic speech recognition (ASR). It is shown that the probability density of a system that consists of a VQ and a discrete classifier can be interpreted as a special case of a semicontinuous mixture model. Thus, the VQ parameters and the classifier can be trained jointly. In this framework, a gradient based VQ training method for single and multiple feature stream systems is derived. This leads to an approach that is directly related to the paradigm of maximum mutual information (MMI) neural networks, that has been successfully applied as VQ in ASR earlier. In continuous speech recognition experiments that were carried out for the Resource Management and Wall Street Journal databases the presented systems achieve recognition accuracies that compete well with comparable Gaussian mixture HMMs. Thus, we demonstrate that the performance degradations, often reported for discrete HMM systems, are not mainly caused by the vector quantization process in itself, but that they are due to the traditional separation of the VQ and the HMM during parameter estimation. These degradations can be avoided by training of the entire system as described here, while keeping the attractive computational speed of discrete HMMs.

Two approaches are proposed for the design of tied-mixture hidden Markov models (TMHMM). One approach improves parameter sharing via partial tying of TMHMM states. To facilitate tying at the substate level, the state emission probabilities are constructed in two stages or, equivalently, are viewed as a "mixture of mixtures of Gaussians." This paradigm allows, and is complemented with, an optimization technique to seek the best complexity-accuracy tradeoff solution, which jointly exploits Gaussian density sharing and substate tying. Another approach to enhance model training is combined training and reduction of model parameters. The procedure starts by training a system with a large universal codebook of Gaussian densities. It then iteratively reduces the size of both the codebook and the mixing coefficient matrix, followed by parameter re-training. The additional cost in design complexity is modest. Experimental results on the ISOLET database and its E-set subset show that substate tying reduces the classification error rate by over 15%, compared to standard Gaussian sharing and whole-state tying. TMHMM design with combined training and reduction of parameters reduces the classification error rate by over 20% compared to conventional TMHMM design. When the two proposed approaches were integrated, 25% error rate reduction over TMHMM with whole-state tying was achieved.

A deterministic annealing algorithm for the design of tiedmixture HMM recognizers is proposed, which reduces the training sensitivity to parameter initialization, automatically smoothes the classification error cost function to allow gradientbased optimization, and seeks better solutions than known techniques. The new approach introduces randomness into the classification rule during the training process, and minimizes the expected error rate while controlling the level of randomness via a constraint on the Shannon entropy. As the entropy constraint is gradually relaxed, the effective cost function converges to the classification error rate and the system becomes a hard (nonrandom) recognizer. Experiments show that the proposed method outperforms design by maximum likelihood reestimation and by generalized probabilistic descent.

Learning-Based Vision and Its Application to Autonomous Indoor Navigation By Shaoyun Chen Adaptation is critical to autonomous navigation of mobile robots. Many adaptive mechanisms have been implemented, ranging from simple color thresholding to complicated learning with artificial neural networks (ANN). The major focus of this thesis lies in machine learning for vision-based navigation. Two well known vision-based navigation systems are ALVINN and ROBIN developed by Carnegie-Mellon University and University of Maryland, respectively. ALVINN uses a two-layer feedforward neural network while ROBIN relies on a radial basis function network (RBFN). Although current ANN-based methods have achieved great success in vision-based navigation, they have two major disadvantages: (1) Local minimum problem: The training of either multilayer perceptron or radial basis function network can get stuck at poor local minimums. (2) The flexibility problem: After the system has been trained in certain r...