Efficient decoding with generative score-spaces using the expectation semiring

State-of-the-art speech recognisers are usually based on hidden Markov models (HMMs). They model a hidden symbol sequence with a Markov process, with the observations independent given that sequence. These assumptions yield efficient algorithms, but limit the power of the model. An alternative model that allows a wide range of features, including word- and phone-level features, is a log-linear model. To handle, for example, word-level variable-length features, the original feature vectors must be segmented into words. Thus, decoding must find the optimal combination of segmentation of the utterance into words and word sequence. Features must therefore be extracted for each possible segment of audio. For many types of features, this becomes slow. In this paper, long-span features are derived from the likelihoods of word HMMs. Derivatives of the log-likelihoods, which break the Markov assumption, are appended. Previously, decoding with this model took cubic time in the length of the sequence, and longer for higher-order derivatives. This paper shows how to decode in quadratic time. © 2013 IEEE.

Distribution of microcystins in various organs (heart, liver, intestine, gonad, brain, kidney and lung) of Wistar rat via intravenous injection

The distribution of microcystins (MCs) in various tissues of Wistar rats was studied under laboratory conditions. Rats were injected intravenously (i.v.) with extracted MCs at a dose of 80 mu g MC-LRequivalent/kg body weight. MCs concentrations in various tissues were detected at 1, 2. 4, 6, 12 and 24 h post-injection using liquid chromatography-mass spectrometry (LC-MS). The highest concentration of MCs was found in kidney (0.034-0.295 mu g/g dry weight), followed by lung (0.007-0.067 mu g/g dry weight), stomach (0.010-0.058 mu g/g dry weight) and liver (0.003-0.052 mu g/g dry weight). The maximum MCs content in the whole body of rat, 2.9% of the injected dose, was observed at 2 h post-injection. MCs concentration was higher in kidney than in liver during the experiment, and two peaks of MCs concentration (at 2 and 24 h, respectively) were observed in kidney, indicating that MCs can be excreted directly via kidney of rat. Though heart, intestine, spleen, brain, gonad and stomach contained less than 0.2% of injected MCs during the whole experiment stage, the presence of MCs in these tissues represents potential damage to them. (c) 2008 Elsevier Ltd. All Fights reserved.

The effect of cyanobacterial crude extract on the transcription of GST mu, GST kappa and GST rho in different organs of goldfish (Carassius auratus)

The glutathione S-transferases play important roles in the detoxification of microcystin. Core-sequences of three classes of GST (mu, kappa and rho) were cloned from goldfish (Carassius auratus L) i.p. injected with cyanobacterial crude extract at two doses (50 and 200 mu g MC-LReq kg(-1) BW). The relative changes of the mRNA abundance in liver, kidney and intestine were analyzed by real-time PCR. The transcription of GST mu was inhibited in intestine at both doses and the transcription of GST kappa was inhibited from 12 to 48 h in kidney at both doses. The decreased transcription of GST rho was detected in all three organs at the high dose. It is suggested that transcription inhibition of GST rho might be significant in MCs toxicity at higher toxin concentration in omnivorous freshwater fish. Alteration in transcription of GSTs stimulated by MCs implicates an increased health risk to fish. (C) 2008 Published by Elsevier B.V.

Bioaccumulation of the hepatotoxic microcystins in various organs of a freshwater snail from a subtropical Chinese lake, Taihu lake, with dense toxic microcystis blooms

In this paper, we describe the seasonal dynamics of three common microcystins (MCs MC-RR, MC-YR, and MC-LR) in the whole body, hepatopancreas, intestine, gonad, foot, remaining tissue, and offspring of a freshwater snail, Bellamya aeruginosa, from Gonghu Bay of Lake Taihu, China, where dense toxic Microcystis blooms occur in the warm seasons. Microcystins were determined by liquid chromatography electrospray ionization mass spectrum. Microcystin (MC-RR + MC-YR + MC-LR) content of the offspring and gonad showed high positive correlation, indicating that microcystins could transfer from adult females to their young with physiological connection. This study is the first to report the presence of microcystins in the offspring of the adult snail. The majority of the toxins were present in the intestine (53.6%) and hepatopancreas (29.9%), whereas other tissues contained only 16.5%. If intestines are excluded, up to 64.3% of the toxin burden was allocated in the hepatopancreas. The microcystin content in the intestine, hepatopancreas, and gonad were correlated with the biomass of Microcystis and intracellular and extracellular toxins. Of the analyzed foot samples, 18.2% were above the tolerable daily microcystin intake recommended by the World Health Organization (WHO) for human consumption. This result indicates that public health warnings regarding human ingestion of snails from Taihu Lake are warranted. In addition, further studies are needed to evaluate the occurrence by Microcystis in relation to spatial and temporal changes in water quality.

Seasonal dynamics of the hepatotoxic microcystins in various organs of four freshwater bivalves from the large eutrophic Lake Taihu of subtropical China and the risk to human consumption

So far, little is known on the distribution of hepatotoxic microcystin (MC) in various organs of bivalves, and there is no study on MC accumulation in bivalves from Chinese waters. Distribution pattern and seasonal dynamics of MC-LR, -YR and -RR in various organs (hepatopancreas, intestine, visceral mass, gill, foot, and rest) of four edible freshwater mussels (Anodonta woodiana, Hyriopsis cumingii, Cristaria plicata, and Lamprotula leai) were studied monthly during Oct. 2003-Sep. 2004 in Lake Taihu with toxic cyanobacterial blooms in the summer. Qualitative and quantitative determinations of MCs in the organs were done by LC-MS and HPLC. The major toxins were present in the hepatopancreas (45.5-55.4%), followed by visceral mass with substantial amount of gonad (27.6-35.5%), whereas gill and foot were the least (1.8-5.1%). The maximum MC contents in the hepatopancreas, intestine, visceral mass, gill, foot, and rest were 38.48, 20.65, 1.70, 0.64, 0.58, and 0.61 mu g/g DW, respectively. There were rather good positive correlation in MC contents between intestines and hepatopancreas of the four bivalves (r = 0.75-0.97, p < 0.05). There appeared to be positive correlations between the maximum MC content in the hepatopancreas and the delta(13)C (r = 0.919) or delta(15)N (r = 0.878) of the foot, indicating that the different MC content in the hepatopancreas might be due to different food ingestion. A glutathione (GSH) conjugate of MC-LR was also detected in the foot sample of C. plicata. Among the foot samples analyzed, 54% were above the provisional WHO tolerable daily intake (TDI) level, and the mean daily intakes from the four bivalves were 8-23.5 times the TDI value when the bivalves are eaten as a whole, suggesting the high risk of consuming bivalves in Lake Taihu. (C) 2005 Wiley Periodicals, Inc.

The role of sense organs in the feeding behaviour of Chinese perch

Experiments were conducted to identify the rules of the individual sense organs in the feeding behaviour of Chinese perch Siniperca chuatsi by determining the consumption of natural food after selective removal or blocking of eyes, lateral lines and olfactory organs, and also by observing the behavioural response to visual, mechanical and chemical stimulation by artificial prey. Chinese perch were able to feed properly on live prey fish when either eyes or lateral lines were intact or functional, but could scarcely feed without these two senses. Chinese perch recognized its prey by vision through the perception of motion and shape, and showed a greater dependence on vision in predation when both visual and mechanical cues were available. Chemical stimulation by natural food could not elicit any feeding response in Chinese perch, and gustation was only important to the fish for the last stage of food discrimination in the oropharyngeal cavity. The sensory basis of Chinese perch in feeding is well adapted to its nocturnal stalking hunting strategy. and also explains its peculiar food habit of accepting live prey fish only and refusing dead prey fish or artificial diets. (C) 1998 The Fisheries Society of the British Isles.