Language model combination and adaptation using weighted finite state transducers

In speech recognition systems language model (LMs) are often constructed by training and combining multiple n-gram models. They can be either used to represent different genres or tasks found in diverse text sources, or capture stochastic properties of different linguistic symbol sequences, for example, syllables and words. Unsupervised LM adaptation may also be used to further improve robustness to varying styles or tasks. When using these techniques, extensive software changes are often required. In this paper an alternative and more general approach based on weighted finite state transducers (WFSTs) is investigated for LM combination and adaptation. As it is entirely based on well-defined WFST operations, minimum change to decoding tools is needed. A wide range of LM combination configurations can be flexibly supported. An efficient on-the-fly WFST decoding algorithm is also proposed. Significant error rate gains of 7.3% relative were obtained on a state-of-the-art broadcast audio recognition task using a history dependently adapted multi-level LM modelling both syllable and word sequences. ©2010 IEEE.

Some observations describing the electrochemical surface annealing of austenitic stainless steel

Cold-worked austenitic stainless steels have been subject to a pulsed electrochemical treatment in fairly concentrated aqueous solutions of sodium nitrite. The electrochemical reactions that occur transform the strain-induced martensite phase, originally formed by the cold work, back to the austenite phase. However, unlike the conventional thermal annealing process, electrochemically induced surface annealing also hardens the surface of the alloy. Because the process causes transformation of the surface martensite, we term it "electrochemical surface annealing", despite the fact that it results in an increase in surface hardness.

Enhanced efficiency of near-UV emitting LEDs for solid-state lighting applications

The performance of a series of near-UV (∼385 nm) emitting LEDs, consisting of high efficiency InGaN/AlInGaN QWs in the active region, was investigated. Significantly reduced roll-over of efficiency at high current density was found compared to InGaN/GaN LEDs emitting at a similar wavelength. The importance of optical cavity effects in flip-chip geometry devices has also been investigated. The light output was enhanced by more than a factor of 2 when the lightemitting region was located at an anti-node position with respect to a high reflectivity current injection mirror. A power of 0.49 mW into a numerical aperture of 0.5 was obtained for a junction area of 50μm in diameter and a current of 30 mA, corresponding to a radiance of 30 W/cm2/str.

Investigation of critical state in melt processed YBa2Cu3O7-δ thick films

The magnetic moment of square planar melt processed YBa2Cu3O7-δ thick films is observed to scale with the cube of the sample width at 4.2 K, suggesting that current flow on the length scale of the film determines its magnetization at this temperature. A well-defined discontinuity in slope in the scaling data at a sample width corresponding to the average grain size (≈2 mm) implies the coexistence of distinct intra- and inter-grain critical current densities of 1.1 × 105Acm-2 and 0.4 × 105Acm-2 at 1 T and 4.2 K. The presence of a critical state in the films at 4.2T is confirmed by removing the central section from a specimen. The observed change in magnetic moment is in excellent agreement with theory for fields greater than ≈2 T. A critical state is not observed at 77 K which suggests that the grains are only weakly coupled at the higher temperature. © 1994.