Continuous speech recognition for the TIMIT database using neural networks

Four types of neural networks which have previously been established for speech recognition and tested on a small, seven-speaker, 100-sentence database are applied to the TIMIT database. The networks are a recurrent network phoneme recognizer, a modified Kanerva model morph recognizer, a compositional representation phoneme-to-word recognizer, and a modified Kanerva model morph-to-word recognizer. The major result is for the recurrent net, giving a phoneme recognition accuracy of 57% from the si and sx sentences. The Kanerva morph recognizer achieves 66.2% accuracy for a small subset of the sa and sx sentences. The results for the word recognizers are incomplete.

Probabilistic fault identification using vibration data and neural networks

Bayesian formulated neural networks are implemented using hybrid Monte Carlo method for probabilistic fault identification in cylindrical shells. Each of the 20 nominally identical cylindrical shells is divided into three substructures. Holes of (12±2) mm in diameter are introduced in each of the substructures and vibration data are measured. Modal properties and the Coordinate Modal Assurance Criterion (COMAC) are utilized to train the two modal-property-neural-networks. These COMAC are calculated by taking the natural-frequency-vector to be an additional mode. Modal energies are calculated by determining the integrals of the real and imaginary components of the frequency response functions over bandwidths of 12% of the natural frequencies. The modal energies and the Coordinate Modal Energy Assurance Criterion (COMEAC) are used to train the two frequency-response-function-neural-networks. The averages of the two sets of trained-networks (COMAC and COMEAC as well as modal properties and modal energies) form two committees of networks. The COMEAC and the COMAC are found to be better identification data than using modal properties and modal energies directly. The committee approach is observed to give lower standard deviations than the individual methods. The main advantage of the Bayesian formulation is that it gives identities of damage and their respective confidence intervals.

Dynamic learning with the EM algorithm for neural networks

In this paper, we derive an EM algorithm for nonlinear state space models. We use it to estimate jointly the neural network weights, the model uncertainty and the noise in the data. In the E-step we apply a forwardbackward Rauch-Tung-Striebel smoother to compute the network weights. For the M-step, we derive expressions to compute the model uncertainty and the measurement noise. We find that the method is intrinsically very powerful, simple and stable.

Numerical simulation of the growth of ZnO nanostructures in a tube furnace by physical vapour deposition

Zinc oxide is a versatile II-VI naturally n-type semiconductor that exhibits piezoelectric properties. By controlling the growth kinetics during a simple carbothermal reduction process a wide range of 1D nanostructures such as nanowires, nanobelts, and nanotetrapods have been synthesized. The driving force: for the nanostructure growth is the Zn vapour supersaturation and supply rate which, if known, can be used to predict and explain the type of crystal structure that results. A model which attempts to determine the Zn vapour concentration as a function of position in the growth furnace is described. A numerical simulation package, COMSOL, was used to simultaneously model the effects of fluid flow, diffusion and heat transfer in a tube furnace made specifically for ZnO nanostructure growth. Parameters such as the temperature, pressure, and flow rate are used as inputs to the model to show the effect that each one has on the Zn concentration profile. An experimental parametric study of ZnO nanostructure growth was also conducted and compared to the model predictions for the Zn concentration in the tube. © 2008 Materials Research Society.