The Minimum Bayes Risk (MBR) framework has been a successful strategy for the training of hidden Markov models for large vocabulary speech recognition. Practical implementations of MBR must select an appropriate hypothesis space and loss function. The set of word sequences and a word-based Levenshtein distance may be assumed to be the optimal choice but use of phoneme-based criteria appears to ...

This study examined the relationship between molecular properties and the fate of trace organic contaminants (TrOCs) in the aqueous and solid phases during wastewater treatment by MBR. A set of 29 TrOCs was selected to represent pharmaceuticals, steroid hormones, phytoestrogens, UV-filters and pesticides that occur ubiquitously in municipal wastewater. Both adsorption and biodegradation/transfo...

One of the major reasons for translation errors in phrase-based SMT systems is the incorrect phrases induced from inaccuracy word-aligned parallel data. In this paper, we propose a novel approach that uses the minimum Bayes-risk (MBR) principle to improve the accuracy of phrase extraction. Our approach performs as a four-stage pipeline: first, bilingual phrases are extracted from parallel corpu...

In this study, effects of organic loading rates on biological nutrient removal and in an MBR based nutrient removal system were investigated. Furthermore, trans-membrane pressure at different applied conditions was monitored to observe impact on membrane fouling Three different levels of CODinf and three HRTs (4, 5 & 8 hrs) were applied. An improved performance with 74 % & 89% of TN & TP remova...

A membrane bioreactor (MBR) was operated for the removal of methyl tert-butyl ether (MTBE) from water. Although the reactor was seeded with several cultures acclimated to MTBE degradation, a long start-up time was observed. Monitoring of the reactor with denaturing gradient gel electrophoresis (DGGE) revealed a dramatic shift in the MBR culture from the original seed culture, indicating that th...

Four rates of methyl bromide (Mbr) (16.8, 33.6, 67,2, and 134.4 kg a.i./ha) and one rate of 1,3-dichloropropene (1,3-D) (28.1 liters a.i./ha) were evaluated over 2 years for control of Hoplolaimus columbus on cotton. All nematicide treatments were applied through a tarpless subsoiler-bedder prior to planting cotton, Gossypium hirsutum cv. Dehapine 90. Nematode population densities were monitore...

A systematic study on the effects of microtaper fiber diameters on the spectral characteristics of a whisperinggallery-mode (WGM) micro-bottle resonator (MBR) is presented. Progressively cleaner and simpler spectra of the MBR were observed when the utilized microtaper fiber waist diameter (Dt) was increased from 2 μm to 10 μm. The maximum transmission depth at resonance varies with different mi...

a bioreactor equipped with hollow fiber microfiltration membranes was applied for wastewater treatment. removal of chemical oxygen demand (cod) and biochemical oxygen demand (bod5) was investigated. the experimental setup consisted of influent and effluent tanks, and membrane modules using polyvinyl di–fluoride (pvdf) hollow fibers. the operation program included suction and backwash steps whic...

The rejection of cyclophosphamide (CP) by nanofiltration (NF) and reverse osmosis (RO) membranes from ultrapure (Milli-Q) water and membrane bioreactor (MBR) effluent was investigated. Lyophilization-extraction and detection methods were first developed for CP analysis in different water matrices. Experimental results showed that the RO membrane provided excellent rejection (>90%) under all ope...

In this paper we describe a method that can be used for Minimum Bayes Risk (MBR) decoding for speech recognition. Our algorithm can take as input either a single lattice, or multiple lattices for system combination. It has similar functionality to the widely used Consensus method, but has a clearer theoretical basis and appears to give better results both for MBR decoding and system combination...