The generation of wear surfaces

Abrasive wear is likely to occur whenever a hard asperity or a trapped hard particle is dragged across a softer surface, and it has been estimated that this form of wear contributes to as many as half of the wear problems that are met in industry. Such damaging hard particles may be external contaminants, products of corrosion or even the debris from previous wear events. During the life of a component, damage caused by individual asperity or particle interactions builds up and, at each stage of its life, the worn surface is the result of many such superimposed wear events. The practical, quantitative prediction of wear rates depends on having both a satisfactory understanding of individual interactions and a suitable procedure for combining these when subsequent contacts are made on a surface whose topography and material properties may have been much changed Irom their initial states. The paper includes some details of an analytical model for the interaction of a representative asperity and the worn surface which can both predict the frictional force and the balance between ploughing, when material is displaced but not lost from the surface, and micromachining or cutting, when actual detachment occurs. Experiments tö !rvvéSuQ8Î8 the validity of the model have been carried out on a novel wear rig which provides very precise control over the position of the asperity and the counterface. This facility, together with that of on-board profilometry, means that it is possible to carry out wear experiments on areas of the surface whose previous deformation history is well known; in this way it is possible to follow the development of a worn surface in a controlled manner as the damage from individual wear events accumulates. Experimental data on the development of such a surface, produced by repeated parallel abrasion, are compared with the predictions of the model. © 1992 IOP Publishing Ltd.

Continuous F0 in the source-excitation generation for HMM-based TTS: Do we need voiced/unvoiced classification?

Most HMM-based TTS systems use a hard voiced/unvoiced classification to produce a discontinuous F0 signal which is used for the generation of the source-excitation. When a mixed source excitation is used, this decision can be based on two different sources of information: the state-specific MSD-prior of the F0 models, and/or the frame-specific features generated by the aperiodicity model. This paper examines the meaning of these variables in the synthesis process, their interaction, and how they affect the perceived quality of the generated speech The results of several perceptual experiments show that when using mixed excitation, subjects consistently prefer samples with very few or no false unvoiced errors, whereas a reduction in the rate of false voiced errors does not produce any perceptual improvement. This suggests that rather than using any form of hard voiced/unvoiced classification, e.g., the MSD-prior, it is better for synthesis to use a continuous F0 signal and rely on the frame-level soft voiced/unvoiced decision of the aperiodicity model. © 2011 IEEE.