Wolfgang Reichl

Decision tree state tying based on penalized Bayesian information criterion

In this paper, an approach of the penalized Bayesian information criterion (pBIC) for decision tree state tying is described. The pBIC is applied to two important applications. First, it is used as a decision tree growing criterion in place of the conventional approach of using a heuristic constant threshold. It is found that original BIC penalty is too low and will not lead to a compact decision tree state tying model. Based on Wolfes modification to the asymptotic null distribution, it is derived that two times BIC penalty should be used for decision tree state tying based on pBIC. Secondly, pBIC is studied as a model compression criterion for decision tree state tying based acoustic modeling. Experimental results on a large vocabulary (Wall Street Journal) speech recognition task indicate that a compact decision tree could be achieved with almost no loss of the speech recognition performance.

A class-based language model for large-vocabulary speech recognition extracted from part-of-speech statistics

A novel approach is presented to class-based language modeling based on part-of-speech statistics. It uses a deterministic word-to-class mapping, which handles words with alternative part-of-speech assignments through the use of ambiguity classes. The predictive power of word-based language models and the generalization capability of class-based language models are combined using both linear interpolation and word-to-class backoff, and both methods are evaluated. Since each word belongs to one precisely ambiguity class, an exact word-to-class backoff model can easily be constructed. Empirical evaluations on large-vocabulary speech-recognition tasks show perplexity improvements and significant reductions in word error-rate.

On natural language call routing

Automated call routing is the process of associating a users request with the desired destination. Although some of the call routing functions can often be accomplished though the use of a touch-tone menu in an interactive voice response system, the interaction between the user and such a system is typically very limited. It is therefore desirable to have a call routing system that takes natural language spoken inputs from the user and asks for additional information to complete the users request as a human agent would. In this paper we present a recent study on natural language call routing and discuss the capabilities and limitations of current technologies.

Discriminative training for continuous speech recognition

Discriminative training techniques for Hidden-Markov Models were recently proposed and successfully applied for automatic speech recognition. In this paper a discussion of the Minimum Classification Error and the Maximum Mutual Information objective is presented. An extended reestimation formula is used for the HMM parameter update for both objective functions. The discriminative training methods were utilized in speaker independent phoneme recognition experiments and improved the phoneme recognition rates for both discriminative training techniques.

An error-protected speech recognition system for wireless communications

Future wireless multimedia terminals will have a variety of applications that require speech recognition capabilities. We consider a robust distributed speech recognition system where representative parameters of the speech signal are extracted at the wireless terminal and transmitted to a centralized automatic speech recognition (ASR) server. We propose two unequal error protection schemes for the ASR bit stream and demonstrate the satisfactory performance of these schemes for typical wireless cellular channels. In addition, a soft-feature error concealment strategy is introduced at the ASR server that uses soft-outputs from the channel decoder to compute the marginal distribution of only the reliable features during likelihood computation at the speech recognizer. This soft-feature error concealment technique reduces the ASR error rate by more than a factor of 2.5 for certain channels. Also considered is a channel decoding technique with source information that improves ASR performance.

A study on natural language call routing

Automated call routing is the process of associating a users request with the desired destination. Although some of the call routing functions can often be accomplished though the use of a touch-tone menu in an interactive voice response system, the interaction between the user and such a system is typically very limited. It is therefore desirable to have a call routing system that takes natural language spoken inputs from the user and asks for additional information to complete the users request as a human agent would. We present a recent study on natural language call routing and discuss the capabilities and limitations of current technologies.

A new model-discriminant training algorithm for hybrid NN-HMM systems

This paper describes a hybrid system for continuous speech recognition consisting of a neural network (NN) and a hidden Markov model (HMM). The system is based on a multilayer perceptron, which approximates the a-posteriori probability of a sequence of states, derived from semi-continuous hidden Markov models. The classification is based on a total score for each hybrid model, attained from a Viterbi search on the state probabilities. Due to the unintended discrimination between the states in each model, a new training algorithm for the hybrid neural networks is presented. The utilized error function approximates the misclassification rate of the hybrid system. The discriminance between the correct and the incorrect models is optimized during the training by the Generalized Probabilistic Descent Algorithm, resulting in a minimum classification error. No explicit target values for the neural net output nodes are used, as in the usual backpropagation algorithm with a quadratic error function. In basic experiments up to 56% recognition rate were achieved on a vowel classification task and up to 69% on a consonant cluster classification task.<<ETX>>

Neural networks for nonlinear discriminant analysis in continuous speech recognition

In this paper neural networks for Nonlinear Discriminant Analysis in continuous speech recognition are presented. Multilayer Perceptrons are used to estimate a-posteriori probabilities for Hidden-Markov Model states, which are the optimal discriminant features for the separation of the HMM states. The a-posteriori probabilities are transformed by a principal component analysis to calculate the new features for semicontinuous HMMs, which are trained by the known Maximum-Likelihood training. The nonlinear discriminant transformation is used in speaker-independent phoneme recognition experiments and compared to the standard Linear Discriminant Analysis technique.