We extend existing methods for automatic sentence boundary detection by leveraging multiple recognizer hypotheses in order to provide robustness to speech recognition errors. For each hypothesized word sequence, an HMM is used to estimate the posterior probability of a sentence boundary at each word boundary. The hypotheses are combined using confusion networks to determine the overall most lik...

English speech based on accent dependent parallel phoneme recognition (PPR) has been developed. The classifier is designed to process continuous speech and to discriminate between native Australian English (AuE) speakers and two migrant speaker groups with foreign accents, whose first languages are Lebanese Arabic (LA) and South Vietnamese (SV). The training of the system can be automated and i...

Why do people turn out to vote? By requiring coordination and by generating positive externalities on others while involving a private cost, turning out to vote resembles a public good contribution and is therefore subject to collective action problems. While this has been established theoretically, the empirical evidence is fraught with measurement and identification problems. We investigate w...

The goal of verification in speech recognition systems is to detect words in the hypothesized sentence that are likely to have been missrecognized. This decision can be based on the persistence of the different words in the output of the speech recognizer when some recognition parameter is varied. To this end, a parameter that proves particularly adequate is the so called Grammar Scale Factor (...

We introduce and analyze a mixture model for supervised learning of probabilistic transducers. We devise an online learning algorithm that efficiently infers the structure and estimates the parameters of each model in the mixture. Theoretical analysis and comparative simulations indicate that the learning algorithm tracks the best model from an arbitrarily large (possibly infinite) pool of mode...

Language modeling for speech recognizer in dialog systems can take two forms. Human input can be constrained through a directed dialog, allowing the decoder to use a state-specific language model to improve recognition accuracy. Mixedinitiative systems allow for human input that while domainspecific might not be state-specific. Nevertheless, for the most part human input to a mixed-initiative s...

Automatic Speech Recognition systems integrate three main knowledge sources: acoustic models, pronunciation dictionary and language models. In contrast to common practices, where each source is optimized independently, then combined in a finite-state search space, we investigate here a training procedure which attempts to adjust (some of) the parameters after, rather than before, combination. T...

In this paper, we present an algorithm for generating test purpose descriptions in form of MSC’s from a given labeled event structure that represents the behavior of a system of asynchronously communicating extended finite state machines. The labeled event structure is a non-interleaving behavior model describing the behavior of a system in terms of the partial ordering of events.

We investigate the combinatorics of a topological space that is generated by the set of edgeweighted finite trees. This space arises by multiplying the weights of edges on paths in trees and is closely connected to tree reconstruction problems involving finite state Markov processes. We show that this space is a contractible finite CW-complex whose face poset can be described via a partial orde...

Formant frequencies have rarely been used as acoustic features for speech recognition, in spite of their phonetic significance. For some speech sounds one or more of the formants may be so badly defined that it is not useful to attempt a frequency measurement. Also, it is often difficult to decide which formant labels to attach to particular spectral peaks. This paper describes a new method of ...