Exploiting variable-width features in large vocabulary speech recognition

The use of variable-width features (prosodics, broad structural information etc.) in large vocabulary speech recognition systems is discussed. Although the value of this sort of information has been recognized in the past, previous approaches have not been widely used in speech systems because either they have not been robust enough for realistic, large vocabulary tasks or they have been limited to certain recognizer architectures. A framework for the use of variable-width features is presented which employs the N-Best algorithm with the features being applied in a post-processing phase. The framework is flexible and widely applicable, giving greater scope for exploitation of the features than previous approaches. Large vocabulary speech recognition experiments using TIMIT show that the application of variable-width features has potential benefits.

Influence of a bow of finite width on bowed string motion: numerical modelling and experimental evidence

An enhanced physical model of the bowed string presented previously [1] is explored. It takes into account: the width of the bow, the angular motion of the string, bow-hair elasticity and string bending stiffness. The results of an analytical investigation of a model system - an infinite string sticking to a bow of finite width and driven on one side of the bow - are compared with experimental results published by Cremer [2] and reinterpreted here. Comparison shows that both the width of the bow and the bow-hair elasticity have a large impact on the reflection and transmission behaviour. In general, bending stiffness plays a minor role. Furthermore, a method of numerical simulation of the stiff string bowed with a bow of finite width is presented along with some preliminary results.

Blade row interaction in a high pressure steam turbine

This paper presents a study of the three-dimensional flow field within the blade rows of a high-pressure axial flow steam turbine stage. Compound lean angles have been employed to achieve relatively low blade loading for hub and tip section and so reduce the secondary losses. The flow field is investigated in a Low-Speed Research Turbine using pneumatic and hot-wire probes downstream of the blade row. Steady and unsteady numerical simulations were performed using structured 3D Navier-Stokes solver to further understand the flow field. Agreement between the simulations and the measurements has been found. The unsteady measurements indicate that there is a significant effect of the stator flow interaction in the downstream rotor blade. The transport of the stator viscous flow through the rotor blade row is described. Unsteady numerical simulations were found to be successful in predicting accurately the flow near the secondary flow interaction regions compared to steady simulations. A method to calculate the unsteady loss generated inside the blade row was developed from the steady numerical simulations. The contribution of various regions in the blade to the unsteady loss generation was evaluated. This method can assist the designer in identifying and optimizing the features of the flow that are responsible for the majority of the unsteady loss production. An analytical model was developed to quantify this effect for the vortex transport inside the downstream blade.

Secondary flows and loss caused by blade row interaction in a turbine stage

A study of the three-dimensional stator-rotor interaction in a turbine stage is presented. Experimental data reveal vortices downstream of the rotor which are stationary in the absolute frame - indicating that they are caused by the stator exit flowfield. Evidence of the rotor hub passage vortices is seen, but additional vortical structures away from the endwalls, which would not be present if the rotor were tested in isolation, are also identified. An unsteady computation of the rotor row is performed using the measured stator exit flowfield as the inlet boundary condition. The strength and location of the vortices at rotor exit are predicted. A formation mechanism is proposed whereby stator wake fluid with steep spanwise gradients of absolute total pressure is responsible for all but one of the rotor exit vortices. This mechanism is then verified computationally using a passive-scalar tracking technique. The predicted loss generation through the rotor row is then presented and a comparison made with a steady calculation where the inlet flow has been mixed out to pitchwise uniformity. The loss produced in the steady simulation, even allowing for the mixing loss at inlet, is 10% less than that produced in the unsteady simulation. This difference highlights the importance of the time-accurate calculation as a tool of the turbomachine designer.

A comparison of losses in small (<1 kW) drives using sine and space vector pulse width modulation schemes

this paper quantifies effects of using three different pulse width modulation (PWM) schemes on the losses in the inverter and induction motor of a 1 kW drive. Direct measurements of losses have been made with a calorimeter. Results show that for the inverter, discontinuous PWM excitation reduces losses by up to 15% compared to sine and symmetrical space vector PWM methods. However, at a low modulation index the greater harmonic content with discontinuous PWM increased motor losses by nearly 20%. This study demonstrates the importance of careful choice of modulation scheme to achieve high overall drive efficiency. © 2005 IEEE.