The use of syntax and multiple alternatives in the VODIS voice operated database inquiry system

The paper describes the architecture of VODIS, a voice operated database inquiry system, and presents some experiments which investigate the effects on performance of varying the level of a priori syntactic constraints. The VODIS system includes a novel mechanism for incorporating context-free grammatical constraints directly into the word recognition algorithm. This allows the degree of a priori constraint to be smoothly varied and provides for the controlled generation of multiple alternatives. The results show that when the spoken input deviates from the predefined task grammar, a combination of weak a priori syntax rules in conjunction with full a posteriori parsing on a lattice of alternative word matches provides the most robust recognition performance. © 1991.

MMI training for continuous phoneme recognition on the TIMIT database

This paper reports our experiences with a phoneme recognition system for the TIMIT database which uses multiple mixture continuous density monophone HMMs trained using MMI. A comprehensive set of results are presented comparing the ML and MMI training criteria for both diagonal and full covariance models. These results using simple monophone HMMs show clear performance gains achieved by MMI training, and are comparable to the best reported by others including those which use context-dependent models. In addition, the paper discusses a number of performance and implementation issues which are crucial to successful MMI training.

The design and implementation of dialogue control in voice operated database inquiry systems

This paper describes work performed as part of the U.K. Alvey sponsored Voice Operated Database Inquiry System (VODIS) project in the area of intelligent dialogue control. The principal aims of the work were to develop a habitable interface for the untrained user; to investigate the degree to which dialogue control can be used to compensate for deficiencies in recognition performance; and to examine the requirements on dialogue control for generating natural speech output. A data-driven methodology is described based on the use of frames in which dialogue topics are organized hierarchically. The concept of a dynamically adjustable scope is introduced to permit adaptation to recognizer performance and the use of historical and hierarchical contexts are described to facilitate the construction of contextually relevant output messages. © 1989.

Stiffness of sands through a laboratory test database

Deformations of sandy soils around geotechnical structures generally involve strains in the range small (0·01%) to medium (0·5%). In this strain range the soil exhibits non-linear stress-strain behaviour, which should be incorporated in any deformation analysis. In order to capture the possible variability in the non-linear behaviour of various sands, a database was constructed including the secant shear modulus degradation curves of 454 tests from the literature. By obtaining a unique S-shaped curve of shear modulus degradation, a modified hyperbolic relationship was fitted. The three curve-fitting parameters are: an elastic threshold strain γe, up to which the elastic shear modulus is effectively constant at G0; a reference strain γr, defined as the shear strain at which the secant modulus has reduced to 0·5G0; and a curvature parameter a, which controls the rate of modulus reduction. The two characteristic strains γe and γr were found to vary with sand type (i.e. uniformity coefficient), soil state (i.e. void ratio, relative density) and mean effective stress. The new empirical expression for shear modulus reduction G/G0 is shown to make predictions that are accurate within a factor of 1·13 for one standard deviation of random error, as determined from 3860 data points. The initial elastic shear modulus, G0, should always be measured if possible, but a new empirical relation is shown to provide estimates within a factor of 1·6 for one standard deviation of random error, as determined from 379 tests. The new expressions for non-linear deformation are easy to apply in practice, and should be useful in the analysis of geotechnical structures under static loading.

Random function priors for exchangeable arrays with applications to graphs and relational data

A fundamental problem in the analysis of structured relational data like graphs, networks, databases, and matrices is to extract a summary of the common structure underlying relations between individual entities. Relational data are typically encoded in the form of arrays; invariance to the ordering of rows and columns corresponds to exchangeable arrays. Results in probability theory due to Aldous, Hoover and Kallenberg show that exchangeable arrays can be represented in terms of a random measurable function which constitutes the natural model parameter in a Bayesian model. We obtain a flexible yet simple Bayesian nonparametric model by placing a Gaussian process prior on the parameter function. Efficient inference utilises elliptical slice sampling combined with a random sparse approximation to the Gaussian process. We demonstrate applications of the model to network data and clarify its relation to models in the literature, several of which emerge as special cases.