Search

Aims The aim was to exploit whole genome sequencing (WGS) to assess genomic diversity, identify virulence genes and deduce the proportion of Campylobacter colonized broilers that directly contaminate their carcasses. Methods and Results Campylobacter jejuni isolates (107) from caeca and carcass neck skin samples (50 pairs from the same batch plus 7 individual caeca) sampled at three poultry slaughterhouses over a one-year period were selected for sequencing (MiSeq; Illumina). FastQ files were submitted to BioNumerics for analysis using the wgMLST scheme for allele calling. Campylobacter cgMLST and hierarchical clustering was performed by applying the single linkage algorithm. Sequence types (STs) were determined in silico from the WGS data and isolates were assigned into clonal complexes (CCs) using the Campylobacter PubMLST.org database. Virulence genes were determined by downloading core sequences from the virulence factor database (VFDB) and the National Center for Biotechnology Information (NCBI). A high degree of diversity was observed with 23 different STs identified. ST257 and CC-21 were the most common STs and CCs, respectively. cgMLST analysis suggested that 56% of carcass contamination was a direct result of contamination from caeca from the same batch. Virulence genes known to play a role in human C. jejuni infection were identified such as the wlaN gene and the genes associated with lipooligosaccharide synthesis, which were identified in 30% of isolates. Conclusions Caecal colonization was the more plausible occurring source of C. jejuni contamination of broiler carcasses, compared with cross-contamination from another batch or the environment. The high rate of genetic diversity observed amongst caecal isolates is consistent with a wide variety of Campylobacter strains circulating in poultry flocks in Ireland. Significance and Impact of Study The results will further inform broiler processors and regulators about the influence and importance of on-farm colonization versus slaughterhouse cross-contamination and the relationship between C. jejuni in caeca and carcasses during processing.

Bovine tuberculosis (bTB) remains a challenging endemic pathogen of cattle in many parts of the globe. Spatial clustering of Mycoacterium bovis molecular types in cattle suggests that local factors are the primary drivers of spread. Northern Ireland’s agricultural landscape is comprised of highly fragmented farms, distributed across spatially discontinuous land parcels, and these highly fragmented farming structures are thought to facilitate localised spread. We conducted a matched case control study to quantify the risks of bTB breakdown with farm area, farm fragmentation, fragment dispersal, and contact with neighbouring herds. Whilst our results show small but significant increases in breakdown risk associated with each of farm fragmentation, farm area, fragment dispersal, and contact with neighbouring herds, these relationships were strongly confounded with the number of contiguous neighbours with bTB. Our key finding was that every infected neighbour led to an increase in the odds of breakdown by 40% to 50%, and that highly fragmented farms were almost twice as likely to have a bTB positive neighbour compared to non-fragmented farms. Our results suggest that after controlling for herd size, herd type, spatial and temporal factors, farm fragmentation increasingly exposes herds to infection originating from first order spatial neighbours. Given NI’s particularly fragmented landscape, and reliance on short-term leases, our data supports the hypothesis that between-herd contiguous spread is a particularly important component of NI’s bTB disease system.

Vocabulary selection, or lexical shortlisting, is a well-known technique to improve latency of Neural Machine Translation models by constraining the set of allowed output words during inference. The chosen set is typically determined by separately trained alignment model parameters, independent of the source-sentence context at inference time. While vocabulary selection appears competitive with respect to automatic quality metrics in prior work, we show that it can fail to select the right set of output words, particularly for semantically non-compositional linguistic phenomena such as idiomatic expressions, leading to reduced translation quality as perceived by humans. Trading off latency for quality by increasing the size of the allowed set is often not an option in real-world scenarios. We propose a model of vocabulary selection, integrated into the neural translation model, that predicts the set of allowed output words from contextualized encoder representations. This restores translation quality of an unconstrained system, as measured by human evaluations on WMT newstest2020 and idiomatic expressions, at an inference latency competitive with alignment-based selection using aggressive thresholds, thereby removing the dependency on separately trained alignment models., Comment: NAACL 2022

Background Salmonella is an important zoonotic pathogen and is one of the main causes of foodborne outbreaks and infections in the European Union. Pigs are a significant reservoir and are frequently subclinical carriers of this organism. Salmonella can be shed in the faeces allowing infection to spread to other pigs, the environment, transport vehicles, lairages and other areas. Inadvertent spillage of gut contents during the slaughter process also leads to contamination. A pig Salmonella control programme has operated in Ireland since 2002 but many local surveys and an EUMS baseline survey in 2008 continued to indicate high levels of the organism in the pig sector. The objectives of this study were to generate updated information on the prevalence of Salmonella spp, in slaughter pigs and carcasses in Irish abattoirs. Five pigs from each of 164 herds were randomly sampled over a 14-week period during 2016. One sample from each of the five pigs of; caecal content, ileo-caecal lymph nodes and carcass swabs (pre-chill) were collected. The five caeca and lymph node samples from each herd were processed as one pool of caecal samples and one pool of lymph node samples, respectively, while the five carcass swabs were tested as individual samples. All isolates were characterised by serotyping and antimicrobial susceptibility. Results In total, 235 Salmonella spp. were isolated from 820 individual carcass swabs, 164 pooled lymph nodes and 164 caecal contents. Salmonella spp. were isolated from 54.3% of the caecal contents and from 31.7% of the ileo-caecal lymph node sample pools. A total of 11.5% of carcass-swab samples yielded Salmonella spp. S. Typhimurium 4,[5],12:i:1,2 or its monophasic variant 4,[5],12:i:-: predominated among isolates from all positive samples; accounting for 73% of lymph nodes, 68% of caecal contents and 56% of carcass swab isolates. S. London and S. Derby were the next most common isolated serotypes. Conclusions These results confirm continuing high levels of Salmonella in fattening pigs in Ireland although reductions in carcass contamination compared to previous surveys were noted. A high prevalence of Salmonella in lymph nodes suggests that it remains a significant problem pre slaughter and a challenge to abattoirs in adhering to process hygiene requirements. The high prevalence of monophasic S. Typhimurim 4,[5],12:i:-: is of serious concern. Therefore, it is important to identify contributing factors in the dissemination of this pathogen in the pork industry in order to minimise the risk of human salmonellosis cases.

The ‘Nature Inspired Creative Design’ cluster brought together people from three main areas: Nature and Biology, Art and Design, and Science and Computing. It aimed to establish a forum for cross-fertilisation between disciplines, within the overall topic of adopting nature inspired approaches to creative design. Its members explored a wide range of subjects, including evolution, growth and development, emergence and self-organisation, robustness, natural structures, and human design behaviour and performance. 1

Slaughterhouse or meat factory surveillance to detect factory lesions (FL) at slaughter is an important part of the bovine tuberculosis (bTB) eradication program in Ireland. The objective of this study was to quantify the effectiveness of Irish slaughterhouses or factories in submitting FL and the proportion of those submitted FL confirmed as being due to bTB in slaughtered cattle, and to identify and quantify the association of risk factors at animal, herd, and factory level with FL submission and confirmation. The data consisted of 6,611,854 animals slaughtered in Irish factories from 2014 to 2018 obtained from the Department of Agriculture, Food and Marine (DAFM), Ireland. Selected risk factors for this study included factory, year and month of slaughter, age, sex, breed, animal movement, ever inconclusive in the standard or severe skin test, herd type, herd size, and bTB history. The association of each risk factor on the FL submission and confirmation risk were analysed with univariable followed by a multivariable logistic regression with herd as random effect. Factories were ranked and compared based on the odds ratio (OR) obtained from the univariable (crude OR) and multivariable (adjusted OR) analysis. The average submission risk of all factories was 20 per 10,000 animals slaughtered, ranging from 1 to 42 per 10,000 animals slaughtered, and the average confirmation risk over all factories was 40.72%, ranging from 0.00 to 61.84%. The odds of submitting and confirming FL as bTB positive were higher in animals over eight years old compared to animals 1–2 years old (OR = 1.91, 95 CI 95% 1.77–2.06 and OR = 4.05, 95% CI 3.17–5.18, respectively), and were higher in animals that ever had inconclusive skin result based on severe interpretation (OR = 2.83, 95% CI 2.44–3.27 and OR = 4.48, 95% CI 2.66–7.54, respectively), animals originating from sucklers herds (OR = 1.08, 95% CI 1.02–1.14 and OR = 1.31, 95% CI 1.11–1.55, respectively), or herds with bTB history in the past three years (OR = 4.46, 95% CI 4.28–4.66 and OR = 319.90, 95% CI 237.98–430.04, respectively). The odds of FL submission and confirmation decreased as the herd size increased (OR = 0.95, 95% CI 0.93–0.96 and OR = 0.82, 95% CI 0.78–0.86, respectively). An inverse relationship of FL submission and confirmation was present for variable sex and inconclusive skin result with standard interpretation, where submission odds were higher in males (OR = 1.05, 95% CI 1.00–1.10) and ever inconclusive animals (OR = 74.24, 95% CI 69.39–79.43), although the confirmation odds were lower (males OR = 0.66, 95% CI 0.56–0.76; ever inconclusive OR = 0.44, 95% CI 0.36–0.54). The crude and adjusted ranking of factories did not differ greatly for FL submission, indicating that factory-related factors may contribute significantly to the submission variation between factories. However, a substantial difference between crude and adjusted confirmation ranking was present which may indicate that animal and herd-related factors were associated to variation in confirmation risk between factories.

Phenotypic antimicrobial resistance testing of 160 Campylobacter jejuni isolates from broiler caeca and neck skin samples collected at key stages during processing from 32 batches was conducted to: (i) determine the prevalence of antimicrobial resistance in C. jejuni isolates (ii) investigate if C. jejuni strains with particular resistance profiles were more frequently recovered at the end of processing, (iii) to establish if multiple strains with different resistance profiles were present within individual batches of broilers and (iv) to determine the virulence profiles of isolates. C. jejuni isolates were tested for resistance to erythromycin (ERY), ciprofloxacin (CIP), tetracycline (TET), gentamicin (GEN), nalidixic acid (NAL) and streptomycin (STR) using a broth dilution phenotypic method. Resistance to TET was most prevalent (46%), followed by resistance to CIP and NAL (29%). All isolates were susceptible to ERY and GEN, while resistance to STR was low (2%). Strains with different resistance profiles were found within 8 individual batches of broilers. Both resistant and susceptible strains were present within 4 individual batches of broilers. Whole genome sequencing of 132 C. jejuni isolates was conducted to: (i) determine relationship between resistant phenotypes and genotypes, (ii) examine genomic diversity of strains present in caeca and in neck skin samples collected at different processing stages and (iii) gain insights into the genomic profiles, including the resistome and virulome of C. jejuni isolates. A high level of correlation between resistant phenotypes and genotypes was observed (99.2%). Resistance determinants identified in sequenced isolates using ResFinder included the tet(O) gene, tet(O/32/O) gene, gyrA p.T86I mutation and blaOXA-61 like genes. A high correlation between sequence type (ST) and antimicrobial resistance was observed. The 132 sequenced C. jejuni isolates were assigned to 10 sequence types (STs), which belonged to 7 clonal complexes, based on multi-locus sequence typing (MLST). ST-257 was prevalent with 58 isolates assigned to it, followed by ST-51 with 25 isolates, ST-10089 with 16 isolates, ST-48 with 13 isolates and ST-50 with 12 isolates. Four isolates were assigned to ST-354, while 1 isolate was assigned to STs 464, 42, 1489 and 6209. All sequenced isolates were assigned to 17 cgSTs. Virulence genes which might have a role in Campylobacter survival and environmental adaptation during key processing stages and a number of virulence genes associated with human infection and colonization of broilers were detected. This study is one of the first to provide data on genomic diversity of C. jejuni isolates from broilers collected during processing in Ireland.

Dialogue State Tracking is central to multi-domain task-oriented dialogue systems, responsible for extracting information from user utterances. We present a novel hybrid architecture that augments GPT-2 with representations derived from Graph Attention Networks in such a way to allow causal, sequential prediction of slot values. The model architecture captures inter-slot relationships and dependencies across domains that otherwise can be lost in sequential prediction. We report improvements in state tracking performance in MultiWOZ 2.0 against a strong GPT-2 baseline and investigate a simplified sparse training scenario in which DST models are trained only on session-level annotations but evaluated at the turn level. We further report detailed analyses to demonstrate the effectiveness of graph models in DST by showing that the proposed graph modules capture inter-slot dependencies and improve the predictions of values that are common to multiple domains., Comment: This paper is in the Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP 2021)

We report on the use of the faecal egg count reduction test to evaluate the performance of ivermectin in treating gastrointestinal nematode infections in first grazing season (FGS) calves on four dairy farms in Co. Kilkenny, Ireland. On each farm, FGS calves were injected subcutaneously with ivermectin in accordance with their live weight (day 0). Calves were individually faecal sampled on both day 0 and day 14. Faecal egg counts were determined using the Mini-FLOTAC technique. Composite faecal cultures for each farm were performed on each sampling occasion. The faecal egg count reductions (mode) ranged from 17.3–80.2% with the lower 95% confidence limit ranging from 3.1–72.3% on the four farms, respectively. Ivermectin-resistant nematodes were detected on all farms, with evidence of Ostertagia resistance on one farm. This study highlights the urgent need for Irish producers to reappraise their parasite control practices. Electronic supplementary material The online version of this article (10.1186/s13620-019-0142-8) contains supplementary material, which is available to authorized users.

This study examined the impact of key processing stages and flock variables on the prevalence of Campylobacter on broiler carcasses. Overall, the prevalence of Campylobacter was 62% in caeca, and 68%, 65% and 62% in neck skin samples collected after evisceration, final wash and carcass chilling, respectively. Campylobacter were found in 32% of caeca, and 52%, 40% and 32% of neck skin samples collected after evisceration, final wash and carcass chilling, respectively from first thin broiler batches. Final thin broiler batches were more frequently contaminated with prevalences of 83% found in caeca, 80% in neck skin samples collected after evisceration and 83% found in neck skin samples collected after both final wash and carcass chilling stages (p < 0.05). Thinning status had a significant effect on Campylobacter counts with significantly higher counts observed in samples from final thin batches (p < 0.05). Highest Campylobacter concentrations in neck skin samples were observed at the evisceration stage in both first and final thin samples, with counts ranging from 2.0 to 3.8 log10 CFU/g and 2.3 to 4.8 log10 CFU/g in first and final thin batches, respectively. All first thin samples had counts below the European Union (EU) Process Hygiene Criterion threshold level of 3 log10 CFU/g after chilling while 52% of final thin batches had counts above this limit.

We examined the pathogens, morphologic patterns, and risk factors associated with bovine respiratory disease (BRD) in 136 recently weaned cattle (“weanlings”), 6–12 mo of age, that were submitted for postmortem examination to regional veterinary laboratories in Ireland. A standardized sampling protocol included routine microbiologic investigations as well as polymerase chain reaction and immunohistochemistry. Lungs with histologic lesions were categorized into 1 of 5 morphologic patterns of pneumonia. Fibrinosuppurative bronchopneumonia (49%) and interstitial pneumonia (48%) were the morphologic patterns recorded most frequently. The various morphologic patterns of pulmonary lesions suggest the involvement of variable combinations of initiating and compounding infectious agents that hindered any simple classification of the etiopathogenesis of the pneumonias. Dual infections were detected in 58% of lungs, with Mannheimia haemolytica and Histophilus somni most frequently recorded in concert. M. haemolytica (43%) was the most frequently detected respiratory pathogen; H. somni was also shown to be frequently implicated in pneumonia in this age group of cattle. Bovine parainfluenza virus 3 (BPIV-3) and Bovine respiratory syncytial virus (16% each) were the viral agents detected most frequently. Potential respiratory pathogens (particularly Pasteurella multocida, BPIV-3, and H. somni) were frequently detected (64%) in lungs that had neither gross nor histologic pulmonary lesions, raising questions regarding their role in the pathogenesis of BRD. The breadth of respiratory pathogens detected in bovine lungs by various detection methods highlights the diagnostic value of parallel analyses in respiratory disease postmortem investigation.

One of the sustainable solutions to the depleting fossils fuels is bioenergy, which is a renewable energy generated from biomass, and biofuel is a hydrocarbon fuel that is produced from biomass. Recently, bioenergy and biofuel projects are encouraged and supported by many governments and organizations in various ways such as providing incentives, technical supports, information, and decision support tools. The economic model is one of the decision support tools, which helps to estimate the costs and earnings involved in a project. It is constructed with various elements such as concepts, relations, logics, constants, and equations. In current economic models, all the elements are hard-coded into some programming code, which makes the model less reusable and extendable. To address the issue, we present an ontology-based economic model in this paper. In particular, we have leveraged the Semantic Web technologies to represent the knowledge about the bioenergy and biofuel economics and inferred the equations and other values required for economic calculations. The case study has been carried out in two of the INTERREG Projects and found promising results.

Minimum Number Alive (MNA) is a widely used index of abundance and trappability in mark-recapture programmes. MNA is a negatively-biased abundance index, and is sensitive to capture probability and the capture session number. This suggests that MNA-trappability will be positively-biased, and that the extent of the bias will vary with true trappability and trapping regime. Inappropriate use of the index may therefore lead to over-optimistic estimates of trappability. In this note, we highlight the dangers of using MNA-trappability for wildlife practitioners. Using closed-population stochastic simulation, we verify that MNA-trappability indeed provides over-optimistic estimates, especially when true trappability is

SUMMARYThe reported incidence of the metastrongylid nematodeAngiostrongylus vasorum, that infects dogs and other canids, is increasing worldwide outside recognized endemic foci. This apparent expansion of the parasite's range is causing concern to veterinary clinicians as the disease caused in dogs can be life threatening and its treatment is not straightforward. The red fox is thought to be a reservoir host for dogs. To investigate the spatial distribution of infection in foxes in Ireland, the hearts and lungs of 542 foxes from all over Ireland were examined. The incidence of infection was found to be 39·9% [95% confidence interval (CI) 35·7–44·1] with positive samples occurring in each of the country's 26 counties. This report confirms that the parasite is endemic in Ireland and the overall prevalence is the second highest in Europe. This is the first survey ofA. vasoruminfection in Irish foxes and highlights the potential exposure of the Irish dog population to high risk of cross-infection. Additionally,Crenosoma vulpiswas found in seven of the foxes, a parasite not previously reported in the Irish fox.

Simple and Efficient Model Filtering in Statistical Machine Translation Data availability and distributed computing techniques have allowed statistical machine translation (SMT) researchers to build larger models. However, decoders need to be able to retrieve information efficiently from these models to be able to translate an input sentence or a set of input sentences. We introduce an easy to implement and general purpose solution to tackle this problem: we store SMT models as a set of key-value pairs in an HFile. We apply this strategy to two specific tasks: test set hierarchical phrase-based rule filtering and n-gram count filtering for language model lattice rescoring. We compare our approach to alternative strategies and show that its trade offs in terms of speed, memory and simplicity are competitive.

In this article we describe HiFST, a lattice-based decoder for hierarchical phrase-based translation and alignment. The decoder is implemented with standard Weighted Finite-State Transducer (WFST) operations as an alternative to the well-known cube pruning procedure. We find that the use of WFSTs rather than k-best lists requires less pruning in translation search, resulting in fewer search errors, better parameter optimization, and improved translation performance. The direct generation of translation lattices in the target language can improve subsequent rescoring procedures, yielding further gains when applying long-span language models and Minimum Bayes Risk decoding. We also provide insights as to how to control the size of the search space defined by hierarchical rules. We show that shallow-n grammars, low-level rule catenation, and other search constraints can help to match the power of the translation system to specific language pairs.

Ivermectin resistant Teladorsagia circumcincta were isolated from two Irish sheep farms. Parasite naive lambs were artificially challenged with the isolates in order to test their sensitivity to benzimidazole (BZ), levamisole (LEV) and ivermectin (IVM) by means of a faecal egg count reduction test (FECRT). Both isolates were found to be resistant to all three anthelmintics with reductions of 47% (C.I. 10-68), 92% (C.I. 74-97) and 50% (C.I. 24-68) for the Farm A isolate and of 85% (C.I. 59-94), 89% (C.I. 69-96) and 73% (C.I. 49-86) for the Farm B isolate for BZ, LEV and IVM, respectively. A controlled efficacy test (CET) was undertaken to confirm resistance to IVM and to determine the efficacy of moxidectin (MOX). In agreement with the FECRT, IVM resistance was confirmed for both isolates, with reductions in worm burden of 78% (C.I. 67-85) and 80% (C.I. 45-93) for the Farm A and B isolates respectively. Both isolates were found to be MOX susceptible, with reductions in worm burden of 97% (C.I. 91-99) and 100% for the Farm A and B isolates respectively. Additionally, in vitro tests confirmed the results of the FECRT; results from the larval development assay (LDA) confirmed that both isolates were resistant to BZ and LEV with development observed at the discriminating dose for each drug, while the larval migration inhibition assay (LMIA) confirmed IVM resistance and indicated MOX susceptibility in both isolates. These results represent the first confirmed cases of IVM resistance on Irish sheep farms as well as the first report of multiple anthelmintic resistant gastrointestinal nematodes in Ireland.

We describe the use of support vector machines (SVMs) for continuous speech recognition by incorporating them in segmental minimum Bayes risk decoding. Lattice cutting is used to convert the Automatic Speech Recognition search space into sequences of smaller recognition problems. SVMs are then trained as discriminative models over each of these problems and used in a rescoring framework. We pose the estimation of a posterior distribution over hypotheses in these regions of acoustic confusion as a logistic regression problem. We also show that GiniSVMs can be used as an approximation technique to estimate the parameters of the logistic regression problem. On a small vocabulary recognition task we show that the use of GiniSVMs can improve the performance of a well trained hidden Markov model system trained under the Maximum Mutual Information criterion. We also find that it is possible to derive reliable confidence scores over the GiniSVM hypotheses and that these can be used to good effect in hypothesis combination. We discuss the problems that we expect to encounter in extending this approach to large vocabulary continuous speech recognition and describe initial investigation of constrained estimation techniques to derive feature spaces for SVMs.

Bioenergy is a renewable energy generated from biomass, while biofuel is a hydrocarbon fuel that is produced from biomass. Recently, bioenergy and biofuel projects are encouraged and supported by many governments and organizations in various ways such as providing incentives, technical supports, information, and decision support tools. Economic model is one of the decision support tools, which helps to estimate the costs and earnings involved in a project. It is constructed with various elements such as concepts, relations, logics, constants and equations. In current economic models, all the elements are hard coded into some programming code, which makes the model less reusable and extendable. To address the issue, we present an ontology-based economic model in this paper. In particular, we have leveraged the Semantic Web technologies to represent the knowledge about the bioenergy and biofuel economics and inferred the equations and other values required for economic calculations. The case study has been carried out in two of the INTERREG Projects and found promising results.

We address the problem of extracting bilingual chunk pairs from parallel text to create training sets for statistical machine translation. We formulate the problem in terms of a stochastic generative process over text translation pairs, and derive two different alignment procedures based on the underlying alignment model. The first procedure is a now-standard dynamic programming alignment model which we use to generate an initial coarse alignment of the parallel text. The second procedure is a divisive clustering parallel text alignment procedure which we use to refine the first-pass alignments. This latter procedure is novel in that it permits the segmentation of the parallel text into sub-sentence units which are allowed to be reordered to improve the chunk alignment. The quality of chunk pairs are measured by the performance of machine translation systems trained from them. We show practical benefits of divisive clustering as well as how system performance can be improved by exploiting portions of the parallel text that otherwise would have to be discarded. We also show that chunk alignment as a first step in word alignment can significantly reduce word alignment error rate.

Minimum Bayes risk estimation and decoding strategies based on lattice segmentation techniques can be used to refine large vocabulary continuous speech recognition systems through the estimation of the parameters of the underlying hidden Markov models and through the identification of smaller recognition tasks which provides the opportunity to incorporate novel modeling and decoding procedures in LVCSR. These techniques are discussed in the context of going 'beyond HMMs', showing in particular that this process of subproblem identification makes it possible to train and apply small-domain binary pattern classifiers, such as Support Vector Machines, to large vocabulary continuous speech recognition.

Lattice segmentation techniques developed for Minimum Bayes Risk decoding in large vocabulary speech recognition tasks are used to compute the statistics needed for discriminative training algorithms that estimate HMM parameters so as to reduce the overall risk over the training data. New estimation procedures are developed and evaluated for both small and large vocabulary recognition tasks, and additive performance improvements are shown relative to maximum mutual information estimation. These relative gains are explained through a detailed analysis of individual word recognition errors.

A framework for dialectal Chinese speech recognition is proposed and studied, in which a relatively small dialectal Chinese (or in other words Chinese influenced by the native dialect) speech corpus and dialect-related knowledge are adopted to transform a standard Chinese (or Putonghua, abbreviated as PTH) speech recognizer into a dialectal Chinese speech recognizer. Two kinds of knowledge sources are explored: one is expert knowledge and the other is a small dialectal Chinese corpus. These knowledge sources provide information at four levels: phonetic level, lexicon level, language level, and acoustic decoder level. This paper takes Wu dialectal Chinese (WDC) as an example target language. The goal is to establish a WDC speech recognizer from an existing PTH speech recognizer based on the Initial-Final structure of the Chinese language and a study of how dialectal Chinese speakers speak Putonghua. The authors propose to use context-independent PTH-IF mappings (where IF means either a Chinese Initial or a Chinese Final), context-independent WDC-IF mappings, and syllable-dependent WDC-IF mappings (obtained from either experts or data), and combine them with the supervised maximum likelihood linear regression (MLLR) acoustic model adaptation method. To reduce the size of the multi-pronunciation lexicon introduced by the IF mappings, which might also enlarge the lexicon confusion and hence lead to the performance degradation, a Multi-Pronunciation Expansion (MPE) method based on the accumulated uni-gram probability (AUP) is proposed. In addition, some commonly used WDC words are selected and added to the lexicon. Compared with the original PTH speech recognizer, the resulting WDC speech recognizer achieves 10–18% absolute Character Error Rate (CER) reduction when recognizing WDC, with only a 0.62% CER increase when recognizing PTH. The proposed framework and methods are expected to work not only for Wu dialectal Chinese but also for other dialectal Chinese languages and even other languages.

We present a Weighted Finite State Transducer Translation Template Model for statistical machine translation. This is a source-channel model of translation inspired by the Alignment Template translation model. The model attempts to overcome the deficiencies of word-to-word translation models by considering phrases rather than words as units of translation. The approach we describe allows us to implement each constituent distribution of the model as a weighted finite state transducer or acceptor. We show that bitext word alignment and translation under the model can be performed with standard finite state machine operations involving these transducers. One of the benefits of using this framework is that it avoids the need to develop specialized search procedures, even for the generation of lattices or N-Best lists of bitext word alignments and translation hypotheses. We report and analyze bitext word alignment and translation performance on the Hansards French-English task and the FBIS Chinese-English task under the Alignment Error Rate, BLEU, NIST and Word Error-Rate metrics. These experiments identify the contribution of each of the model components to different aspects of alignment and translation performance. We finally discuss translation performance with large bitext training sets on the NIST 2004 Chinese-English and Arabic-English MT tasks.

Linear transforms have been used extensively for training and adaptation of HMM-based ASR systems. Recently procedures have been developed for the estimation of linear transforms under the Maximum Mutual Information (MMI) criterion. In this paper we introduce discriminative training procedures that employ linear transforms for feature normalization and for speaker adaptive training. We integrate these discriminative linear transforms into MMI estimation of HMM parameters for improvement of large vocabulary conversational speech recognition systems.

Knowledge acquisition is becoming an integral part of the manufacturing industries, which rely on domain experts in various phases of product life cycle including design, analysis, manufacturing, operation and maintenance. It has the potential to enable knowledge reuse, however, poorly managed knowledge can cause information loss and inefficiency. If technical documentation is managed well in the manufacturing industries, intended piece of knowledge can easily be located, used and reused for purpose and as a result, the corresponding industry can be benefited. Some examples of technical documentation are design specification, operating manual and maintenance manual. Model-based Documentation (MBD) is a documentation approach that uses model to provide structure to the data of the documents. MBD can be thought of as a way to better organize knowledge thereby knowledge identification and retrieval become easier, faster and efficient. In this paper, we propose MBD and its extension as a potential solution to overcome the issues involved in the typical technical documentation approaches.

Bio energy is a renewable energy and a solution to the depleting fossil fuels. Bio energy such as heat, power and bio fuel is generated by conversion technologies using biomass for example domestic waste, root crops, forest residue and animal slurry. Pyrolysis, anaerobic digestion and combined heat and power engine are some examples of the technologies. Depending on the nature of a biomass, it can be treated with various technologies giving out some products, which can be further treated with other technologies and eventually converted into the final products as bio energy. The pathway followed by the biomass, technologies, intermediate products and bio energy in the conversion process is referred to as bio energy pathway. Identification of appropriate pathways optimizes the conversion process. Although there are various approaches to create or generate the pathways, there is still a need for a semantic approach to generate the pathways, which allow checking the consistency of the knowledge, and to share and extend the knowledge efficiently. This paper presents an ontology-based approach to automatic generation of the pathways for biomass to bio energy conversion, which exploits the definition and hierarchical structure of the biomass and technologies, their relationship and associated properties, and infers appropriate pathways. A case study has been carried out in a real-life scenario, the bio energy project for the North West of Europe (Bioen NW), which showed promising results.

Filled pauses are pervasive in conversational speech and have been shown to serve several psychological and structural purposes. Despite this, they are seldom modelled overtly by stateof-the-art speech synthesis systems. This paper seeks to motivate the incorporation of filled pauses into speech synthesis systems by exploring their use in conversational speech, and by comparing the performance of several automatic systems inserting filled pauses into fluent text. Two initial experiments are described which seek to determine whether people’s predicted insertion points are consistent with actual practice and/or with each other. The experiments also investigate whether there are ‘right’ and ‘wrong’ places to insert filled pauses. The results show good consistency between people’s predictions of usage and their actual practice, as well as a perceptual preference for the ‘right’ placement. The third experiment contrasts the performance of several automatic systems that insert filled pauses into fluent sentences. The best performance (determined by F-score) was achieved through the by-word interpolation of probabilities predicted by Recurrent Neural Network and 4gram Language Models. The results offer insights into the use and perception of filled pauses by humans, and how automatic systems can be used to predict their locations. Index Terms: filled pause, HMM TTS, SVM, RNN

The widely used maximum likelihood linear regression speaker adaptation procedure suffers from overtraining when used for rapid adaptation tasks in which the amount of adaptation data is severely limited. This is a well known difficulty associated with the expectation maximization algorithm. We use an information geometric analysis of the expectation maximization algorithm as an alternating minimization of a Kullback–Leibler-type divergence to see the cause of this difficulty, and propose a more robust discounted likelihood estimation procedure. This gives rise to a discounted likelihood linear regression procedure, which is a variant of maximum likelihood linear regression suited for small adaptation sets. Our procedure is evaluated on an unsupervised rapid adaptation task defined on the Switchboard conversational telephone speech corpus, where our proposed procedure improves word error rate by 1.6% (absolute) with as little as 5 seconds of adaptation data, which is a situation in which maximum likelihood linear regression overtrains in the first iteration of adaptation. We compare several realizations of discounted likelihood linear regression with maximum likelihood linear regression and other simple maximum likelihood linear regression variants, and discuss issues that arise in implementing our discounted likelihood procedures.

The paper entitled "Efficient training algorithms for HMMs using incremental estimation" by Gotoh et al. (IEEE Trans. Speech Audio Processing, vol.6, p.539-48, Nov. 1998) investigated expectation maximization (EM) procedures that increase training speed. The claim of Gotoh et al. that these procedures are generalized EM (Dempster et al. 1977) procedures is shown to be incorrect in the present paper. We discuss why this is so, provide an example of nonmonotonic convergence to a local maximum in likelihood, and outline conditions that guarantee such convergence.

In the early 1990s, the availability of the TIMIT read-speech phonetically transcribed corpus led to work at AT&T on the automatic inference of pronunciation variation. This work, briefly summarized here, used stochastic decision trees trained on phonetic and linguistic features, and was applied to the DARPA North American Business News read-speech ASR task. More recently, the ICSI spontaneous-speech phonetically transcribed corpus was collected at the behest of the 1996 and 1997 LVCSR Summer Workshops held at Johns Hopkins University. A 1997 workshop (WS97) group focused on pronunciation inference from this corpus for application to the DoD Switchboard spontaneous telephone speech ASR task. We describe several approaches taken there. These include (1) one analogous to the AT&T approach, (2) one, inspired by work at WS96 and CMU, that involved adding pronunciation variants of a sequence of one or more words (`multiwords') in the corpus (with corpus-derived probabilities) into the ASR lexicon, and (1+2) a hybrid approach in which a decision-tree model was used to automatically phonetically transcribe a much larger speech corpus than ICSI and then the multiword approach was used to construct an ASR recognition pronunciation lexicon. Anfang der neunziger Jahre hat die Verfugbarkeit des phonetisch umschrifteten TIMIT Korpus fur gelesene Sprache bei AT&T zur Forschung uber die automatische Inferenz von Aussprachemodellen gefuhrt. In dieser Arbeit wurden Decision Trees auf der Basis von phonetischen wie linguistischen Merkmalen trainiert und auf den DARPA North American Business News Korpus fur die automatische Erkennung gelesener Sprache angewendet. In jungerer Zeit wurde fur die an der Johns Hopkins Universitat ausgerichteten LVCSR Summer Workshops 1996 und 1997 der phonetisch transkribierte ICSI Korpus fur spontane Sprache erstellt. Eine Arbeitsgruppe des Workshops 1997 (WS97) hat sich auf Basis dieses Korpus mit der Inferenz von Aussprachemodellen zur Anwendung auf den DoD Switchboard Korpus fur spontane Telefonsprache befa?st. Wir beschreiben im folgenden verschiedene Ansatze in diese Richtung. Sie umfassen im einzelnen: ein zu dem AT&T Ansatz (1) analoges Modell, einen Ansatz (2), der auf der Arbeit am WS96 und an der CMU aufbaut und die automatische Generierung und Aufnahme von Aussprachevarianten fur Folgen von einem oder mehreren Worten (`multiwords') in das ASR Ausspracheworterbuch (mit aus dem Trainingskorpus abgeleiteten Wahrscheinlichkeiten) behandelt, und ein hybrides Modell (1+2), in dem ein Decision Tree basiertes Modell zur automatischen phonetischen Transkription eines sehr viel gro?seren als des ICSI Korpus verwendet wird um dann mittels des `multiword'-Ansatzes ein ASR Ausspracheworterbuch zu generieren. Au debut des annees 90, la disponibilite du corpus TIMIT de transcription phonetique de parole dictee conduisit les chercheurs de AT&T a travailler sur l'inference automatique des variations de prononciation. Ce travail, brievement resume ici, utilisa des arbres de decision construit a partir de traits phonetiques et linguistiques, et fut applique a la reconaissance de la parole dictee dans le domaine du DARPA North American Business News. Plus recemment, le corpus ICSI de transcription phonetique de parole spontanee fut mis au point a l'occasion des ateliers de travails d'ete a l'universite Johns Hopkins (WS) en 1996 et 1997. Un groupe de travail du WS97 mit l'accent sur l'inference de la prononciation a partir de ce corpus afin de l'utiliser dans le domaine de la parole spontanee telephonique DoD Switchboard. Nous decrivons plusieurs approches suivies a cette occasion. Ces approches incluent (1) une approche analogue a celle de AT&T, (2) une autre approche inspiree par le travail accompli a WS96 et a CMU qui necessite l'addition de variantes de prononciation d'une sequence d'un ou plusieurs mots du corpus (`multimots', avec des probabilites extraites de ce corpus) au dictionnaire de prononciation du systeme de reconnaissance de la parole, et (1+2) une approche hybride ou un model d'arbres de decision est utilise pour determiner automatiquement la transcription phonetique d'un corpus bien plus long que celui d'ICSI, et ou l'approche multimots est utilisee pour construire un dictionnaire de prononciation pour la reconnaissance automatique de la parole.

We conducted a randomized, double-blind, placebo-controlled multicenter trial of azithromycin (1,200 mg once weekly) for the prevention of Mycobacterium avium complex (MAC) infection in patients with AIDS and a CD4 cell count of < 100/mm3. In an intent-to-treat analysis through the end of therapy plus 30 days, nine (10.6%) of 85 azithromycin recipients and 22 (24.7%) of 89 placebo recipients developed MAC infection (hazard ratio, 0.34; P = .004). There was no difference in the ranges of minimal inhibitory concentrations of either clarithromycin or azithromycin for the five breakthrough (first) MAC isolates from the azithromycin group and the 18 breakthrough MAC isolates from the placebo group. Of the 76 patients who died during the study, four (10.5%) of 38 azithromycin recipients and 12 (31.6%) of 38 placebo recipients had a MAC infection followed by death (P = .025). For deaths due to all causes, there was no difference in time to death or number of deaths between the two groups. Episodes of non-MAC bacterial infection per 100 patient years occurred in 43 azithromycin recipients and 88 placebo recipients (relative risk, 0.49; 95% confidence interval, 0.33-0.73). The most common toxic effect noted during the study was gastrointestinal, reported by 78.9% of azithromycin recipients and 27.5% of placebo recipients. Azithromycin given once weekly is safe and effective in preventing disseminated MAC infection, death due to MAC infection, and respiratory tract infections in patients with AIDS and CD4 cell counts of < 100/mm3.

This paper introduces the Next-TELL independent open learner model which is constructed based on data from a range of sources. An example is presented for a university course, with the learner model built from the main activities undertaken during the course. Use of the Next-TELL open learner model over a five week period is described for this group of students, suggesting that independent open learner models built from multiple sources of data may have much to offer in supporting students’ understanding of their learning, and could potentially be used to encourage greater peer interaction.

This paper draws on visualisation approaches in learning analytics, considering how classroom visualisations can come together in practice. We suggest an open learner model in situations where many tools and activity visualisations produce more visual information than can be readily interpreted.

The Suburban Birmingham project aimed to allow the exploration of a collection of resources originally presented through a website on a touch table. In this paper we will explore some of the issues illustrated by this project involving the manipulation and sharing of resources on multi-user applications in public spaces, including ownership, affordances, and managing and interpreting gestures in different contexts.

OBJECTIVE To determine the prevalence of incipient and overt nephropathy in African- American subjects with non-insulin-dependent diabetes mellitus(NIDDM) attending a hospital clinic. Contributory factors, such as blood pressure (BP), duration and age at onset of diabetes, hyperglycemia, hyperlipidemia, and body mass index (BM1) also were evaluated. RESEARCH DESIGN AND METHODS We recruited 116 African-American subjects with NIDDM for this cross-sectional, descriptive, and analytical study. BP, BMI, 24-h urine albumin excretion, creatinine clearance, serum creatinine, lipids, and GHb levels were measured. Albumin excretion rate (AER) was calculated, and subjects were divided into three groups: no nephropathy (AER 200 μg/min). Frequency of hypertension and nephropathy was analyzed by x2 testing, group means were comparedusing analysis of variance, and linear correlations were performed between AER and other variables. Multiple regression analysis was used to examine theassociation of these variables while controlling for the effects of other variables. RESULTS Increased AER was present in 50% of our subjects; 31% had incipient and 19% had overt nephropathy. Hypertension was present in 72.4%; nephropathy, particularly overt nephropathy, was significantly more prevalent in the hypertensive group. Mean BP and diastolic blood pressure (dBP) were higher in the groups with incipient and overt nephropathy, and systolic blood pressure (sBP) was increased in overt nephropathy. Men with either form of nephropathy had higher sBP, dBP, and mean BP, whereas only women with overt nephropathy had increased sBP and mean BP. Subjects with incipient or overt nephropathy had a longer duration of diabetes, and those with overt nephropathyhad a younger age at onset of diabetes. By multiple regression analysis, AER correlated with younger age at diabetes onset, but not with diabetes duration. No correlation with age, lipid levels, or GHb was noted. BMI correlated with AER. CONCLUSIONS Incipient and overt nephropathy were observed frequently inthese African-American subjects with NIDDM. Albuminuria correlated with BP, younger age at diabetes onset, and BMI. Association of albuminuria and increased cardiovascular mortality may place 50% of inner-city African-American patients with NIDDM at risk for developing cardiovascular complications.

This paper describes a speech recognition system for the credit card corpus which was provided as a baseline for the Summer Workshop on Robust Speech Processing held at the Rutgers CAIP Center in July/August 1993. The system was built using a portable HMM toolkit called HTK. This is described, along with details of the training and testing methods used. Benchmark performance results are presented for mixture Gaussian monophones and tied-state triphones using four different language models. The overall accuracy levels achieved were very low, indicating that this is a very difficult recognition task. The paper concludes with a discussion of some specific problems encountered with the Credit Card data and suggestions for future work. >

In a cooperative mixed-initiative system, timely and effective dialogue between the system and user is important to ensure that both sides work towards producing the most effective results, and this is affected by how disruptive any interruptions are as the user completes their primary task. A disruptive interaction means the user may become irritated with the system, or might take longer to deal with the interruption and provide information that the system needs to continue. Disruption is influenced both by the nature of the interaction and when it takes place in the context of the user’s progress through their main task. We describe an experiment based on a prototype cooperative video annotation system designed to explore the impact of interruptions, in the form of questions posed by the system that the user must address. Our findings demonstrate a preference towards questions presented in context with the content of the video, rather than at the natural opportunities presented by transitions in the video. This differs from previous research which concentrates on interruptions in the form of notifications.

The autoregressive HMM has been shown to provide efficient parameter estimation and high-quality synthesis, but in previous experiments decision trees derived from a non-autoregressive system were used. In this paper we investigate the use of autoregressive clustering for autoregressive HMM-based speech synthesis. We describe decision tree clustering for the autoregressive HMM and highlight differences to the standard clustering procedure. Subjective listening evaluation results suggest that autoregressive clustering improves the naturalness of the resulting speech. We find that the standard minimum description length (MDL) criterion for selecting model complexity is inappropriate for the autoregressive HMM. Investigating the effect of model complexity on naturalness, we find that a large degree of overfitting is tolerated without a substantial decrease in naturalness.

Hidden Markov model (HMM)-based speech synthesis systems possess several advantages over concatenative synthesis systems. One such advantage is the relative ease with which HMM-based systems are adapted to speakers not present in the training dataset. Speaker adaptation methods used in the field of HMM-based automatic speech recognition (ASR) are adopted for this task. In the case of unsupervised speaker adaptation, previous work has used a supplementary set of acoustic models to estimate the transcription of the adaptation data. This paper firstly presents an approach to the unsupervised speaker adaptation task for HMM-based speech synthesis models which avoids the need for such supplementary acoustic models. This is achieved by defining a mapping between HMM-based synthesis models and ASR-style models, via a two-pass decision tree construction process. Secondly, it is shown that this mapping also enables unsupervised adaptation of HMM-based speech synthesis models without the need to perform linguistic analysis of the estimated transcription of the adaptation data. Thirdly, this paper demonstrates how this technique lends itself to the task of unsupervised cross-lingual adaptation of HMM-based speech synthesis models, and explains the advantages of such an approach. Finally, listener evaluations reveal that the proposed unsupervised adaptation methods deliver performance approaching that of supervised adaptation.

The goal of Morpho Challenge 2009 was to evaluate unsupervised algorithms that provide morpheme analyses for words in different languages and in various practical applications. Morpheme analysis is particularly useful in speech recognition, information retrieval and machine translation for morphologically rich languages where the amount of different word forms is very large. The evaluations consisted of: 1. a comparison to grammatical morphemes, 2. using morphemes instead of words in information retrieval tasks, and 3. combining morpheme and word based systems in statistical machine translation tasks. The evaluation languages were: Finnish, Turkish, German, English and Arabic. This paper describes the tasks, evaluation methods, and obtained results. The Morpho Challenge was part of the EU Network of Excellence PASCAL Challenge Program and organized in collaboration with CLEF.

We introduce alignment models for Machine Translation that take into account the context of a source word when determining its translation. Since the use of these contexts alone causes data sparsity problems, we develop a decision tree algorithm for clustering the contexts based on optimisation of the EM auxiliary function. We show that our context-dependent models lead to an improvement in alignment quality, and an increase in translation quality when the alignments are used in Arabic-English and Chinese-English translation.

This paper describes a lattice-based decoder for hierarchical phrase-based translation. The decoder is implemented with standard WFST operations as an alternative to the well-known cube pruning procedure. We find that the use of WFSTs rather than k-best lists requires less pruning in translation search, resulting in fewer search errors, direct generation of translation lattices in the target language, better parameter optimization, and improved translation performance when rescoring with long-span language models and MBR decoding. We report translation experiments for the Arabic-to-English and Chinese-to-English NIST translation tasks and contrast the WFST-based hierarchical decoder with hierarchical translation under cube pruning.