System combination and score normalization for spoken term detection

Spoken content in languages of emerging importance needs to be searchable to provide access to the underlying information. In this paper, we investigate the problem of extending data fusion methodologies from Information Retrieval for Spoken Term Detection on low-resource languages in the framework of the IARPA Babel program. We describe a number of alternative methods improving keyword search performance. We apply these methods to Cantonese, a language that presents some new issues in terms of reduced resources and shorter query lengths. First, we show score normalization methodology that improves in average by 20% keyword search performance. Second, we show that properly combining the outputs of diverse ASR systems performs 14% better than the best normalized ASR system. © 2013 IEEE.

Detection and quantitative analysis of polychlorinated biphenyls in tilapia from Hawaiian waters

The purpose of this study was to investigate polychlorinated biphenyls (PCBs) contamination in tilapia (Oreochromis mossambicus) collected from the Manna stream and Ala Wai Canal of O'ahu, an island of the geographically isolated Hawaiian archipelago. Our results show that the average concentrations of PCBs varied from 51.90 to 89.42 ng g(-1) lipid weight for the sampling sites. Relative toxic potencies (RTPs) and toxic equivalencies (TEQs) were determined to be 20.38-40.60 ng TCDD g(-1) lipid weight and 2.89-4.17 ng TEQ g(-1) lipid weight by 7-ethoxy-resorufin-O-deethylase (EROD) activity analysis and calculation of PCB concentrations based on toxic equivalency factors (TEFs), respectively. Penta-chlorinated congeners were found to be predominant, which revealed that Aroclor 1254 was a possible major source of PCBs in our fish samples. PCB 118, an indicator PCBs, constituted more than 55% and 30% of the total PCBs and TEQs, respectively. In addition, PCB 118 was found to have a linear correlation to the total PCBs (R = 0.975) and TEQs (R = 0.782). Detection of concentrated PCBs in Hawaiian waters suggests a potentially adverse impact of this pollutant on human health, as well as ecological systems, and suggests the necessity of environmental monitoring and hazard assessment of PCBs within the Hawaiian Islands. (c) 2008 Published by Elsevier Ltd.

Investigation of multilingual deep neural networks for spoken term detection

The development of high-performance speech processing systems for low-resource languages is a challenging area. One approach to address the lack of resources is to make use of data from multiple languages. A popular direction in recent years is to use bottleneck features, or hybrid systems, trained on multilingual data for speech-to-text (STT) systems. This paper presents an investigation into the application of these multilingual approaches to spoken term detection. Experiments were run using the IARPA Babel limited language pack corpora (∼10 hours/language) with 4 languages for initial multilingual system development and an additional held-out target language. STT gains achieved through using multilingual bottleneck features in a Tandem configuration are shown to also apply to keyword search (KWS). Further improvements in both STT and KWS were observed by incorporating language questions into the Tandem GMM-HMM decision trees for the training set languages. Adapted hybrid systems performed slightly worse on average than the adapted Tandem systems. A language independent acoustic model test on the target language showed that retraining or adapting of the acoustic models to the target language is currently minimally needed to achieve reasonable performance. © 2013 IEEE.

Estimation of the detection probability for Yangtze finless porpoises (Neophocaena phocaenoides asiaeorientalis) with a passive acoustic method

Yangtze finless porpoises were surveyed by using simultaneous visual and acoustical methods from 6 November to 13 December 2006. Two research vessels towed stereo acoustic data loggers, which were used to store the intensity and sound source direction of the high frequency sonar signals produced by finless porpoises at detection ranges up to 300 m on each side of the vessel. Simple stereo beam forming allowed the separation of distinct biosonar sound source, which enabled us to count the number of vocalizing porpoises. Acoustically, 204 porpoises were detected from one vessel and 199 from the other vessel in the same section of the Yangtze River. Visually, 163 and 162 porpoises were detected from two vessels within 300 m of the vessel track. The calculated detection probability using acoustic method was approximately twice that for visual detection for each vessel. The difference in detection probabilities between the two methods was caused by the large number of single individuals that were missed by visual observers. However, the sizes of large groups were underestimated by using the acoustic methods. Acoustic and visual observations complemented each other in the accurate detection of porpoises. The use of simple, relatively inexpensive acoustic monitoring systems should enhance population surveys of free-ranging, echolocating odontocetes. (C) 2008 Acoustical Society of America.

Wideband wireless over fiber systems

Broadband radio over fiber systems, which can simultaneously distribute multiple wireless services and enable remote sensing, are reviewed. The systems are used to demonstrate improved remote passive RFID tag detection through the use of multiple antennas. © 2009 Optical Society of America.

Direct detection of the relative strength of Rashba and Dresselhaus spin-orbit interaction: Utilizing the SU(2) symmetry

We propose a simple method to detect the relative strength of Rashba and Dresselhaus spin-orbit interactions in quantum wells (QWs) without relying on the directional-dependent physical quantities. This method utilizes the two different critical gate voltages that leading to the remarkable signals of SU(2) symmetry, which happens to reflect the intrinsic-structure-inversion asymmetry of the QW. We support our proposal by the numerical calculation of in-plane relaxation times based on the self-consistent eight-band Kane model. We find that the two different critical gate voltages leading to the maximum spin-relaxation times [one effect of the SU(2) symmetry] can simply determine the ratio of the coefficients of Rashba and Dresselhaus terms. Our proposal can also be generalized to extract the relative strengths of the spin-orbit interactions in quantum-wire and quantum-dot structures.

Ground target detection, classification and sensor fusion in distributed fiber seismic sensor network - art. no. 683015

This paper describes the ground target detection, classification and sensor fusion problems in distributed fiber seismic sensor network. Compared with conventional piezoelectric seismic sensor used in UGS, fiber optic sensor has advantages of high sensitivity and resistance to electromagnetic disturbance. We have developed a fiber seismic sensor network for target detection and classification. However, ground target recognition based on seismic sensor is a very challenging problem because of the non-stationary characteristic of seismic signal and complicated real life application environment. To solve these difficulties, we study robust feature extraction and classification algorithms adapted to fiber sensor network. An united multi-feature (UMF) method is used. An adaptive threshold detection algorithm is proposed to minimize the false alarm rate. Three kinds of targets comprise personnel, wheeled vehicle and tracked vehicle are concerned in the system. The classification simulation result shows that the SVM classifier outperforms the GMM and BPNN. The sensor fusion method based on D-S evidence theory is discussed to fully utilize information of fiber sensor array and improve overall performance of the system. A field experiment is organized to test the performance of fiber sensor network and gather real signal of targets for classification testing.

Research of Face Detection System Applied for Detecting Faces with Complex Background and Illumination

An effective face detection system used for detecting multi pose frontal face in gray images is presented. Image preprocessing approaches are applied to reduce the influence of the complex illumination. Eye-analog pairing and improved multiple related template matching are used to glancing and accurate face detecting, respectively. To shorten the time cost of detecting process, we employ prejudge rules in checking candidate image segments before template matching. Test by our own face database with complicated illumination and background, the system has high calculation speed and illumination independency, and obtains good experimental results.

A colorimetric assay for screening microcystin class compounds in aquatic systems

Secondary metabolites produced by water-blooming cyanobacteria in eutrophic waters include some potent hepatotoxins, These compounds also have tumour-promoting properties, attributable to their inhibition and activation of protein phosphatases and kinases respectively. The inhibitory effect of these toxins on protein phosphatases have been employed in a commonly used radiometric assay, involving the use of a P-32-labeled substrate, for the detection and quantitation of these compounds. This paper investigates and describes a colorimetric method in which the activity of protein phosphatase 2A is determined by measuring the rate of colour production from the release of yellow p-nitrophenol using p-nitrophenyl phosphate as the substrate. Results of this study suggest that the colorimetric protein phosphatase inhibition assay is a simple, inexpensive tool for screening substances that may have tumour-promoting characteristics in aquatic systems. The detection limit of the colorimetric method is comparable to the radiometric assay. (C) 1998 Elsevier Science Ltd. All rights reserved.

VLSI architecture of a K-best detector for MIMO--OFDM wireless communication systems

The K-best detector is considered as a promising technique in the MIMO-OFDM detection because of its good performance and low complexity. In this paper, a new K-best VLSI architecture is presented. In the proposed architecture, the metric computation units (MCUs) expand each surviving path only to its partial branches, based on the novel expansion scheme, which can predetermine the branches' ascending order by their local distances. Then a distributed sorter sorts out the new K surviving paths from the expanded branches in pipelines. Compared to the conventional K-best scheme, the proposed architecture can approximately reduce fundamental operations by 50% and 75% for the 16-QAM and the 64-QAM cases, respectively, and, consequently, lower the demand on the hardware resource significantly. Simulation results prove that the proposed architecture can achieve a performance very similar to conventional K-best detectors. Hence, it is an efficient solution to the K-best detector's VLSI implementation for high-throughput MIMO-OFDM systems.

Sensitive, Label-Free Protein Assay Using 1-ethyl-3-methylimidazolium Tetrafluoroborate Supported Microchip Electrophoresis with Laser-Induced Fluorescence Detection

Based on the dimer-monomer equilibrium movement of the fluorescent dye Pyronin Y (PY), a rapid, simple, highly sensitive, label-free method for protein detection was developed by microchip electrophoresis with LIF detection. PY formed a nonfluorescent dimer induced by the premicellar aggregation of an anionic surfactant, SDS, however, the fluorescence intensity of the system increased dramatically when proteins such as BSA, bovine hemoglobin, cytochrome c, and trypsin were added to the solution due to the transition of dimer to fluorescent monomer. Furthermore, 1-ethyl-3-methylimidazolium tetrafluoroborate (EMImBF(4)) instead of PBS was applied as running buffers in microchip electrophoresis.

Hydrogen-Bonding Recognition-Induced Color Change of Gold Nanoparticles for Visual Detection of Melamine in Raw Milk and Infant Formula

The color change induced by triple hydrogen-bonding recognition between melamine and a cyanuric acid derivative grafted on the surface of gold nanoparticles can be used for reliable detection of melamine. Since such a color change can be readily seen by the naked eye, the method enables on-site and real-time detection of melamine in raw milk and infant formula even at a concentration as low as 2.5 ppb without the aid of any advanced instruments.

Glutathione-capped CdTe quantum dots for the sensitive detection of glucose

A simple and sensitive assay system for glucose based on the glutathione (GSH)-capped CdTe quantum dots (QDs) was developed. GSH-capped CdTe QDs exhibit higher sensitivity to H2O2 produced from the glucose oxidase catalyzed oxidation Of glucose, and are also more biocompatible than other thiols-capped QDs. Based on the quenching of H2O2 on GSH-capped QDs, glucose can be detected. The detection conditions containing reaction time, the concentration of glucose oxidase and the sizes of QDs were optimized and the detection limits for glucose was determined to be 0.1 mu M; two detection ranges of glucose from 1.0 mu M to 0.5 mM and from 1.0 mM to 20 mM, respectively Were obtained. The detection limit was almost a 1000 times lower than other QDs-based optical glucose sensing systems. The developed glucose detection system was simple and facile with no need of complicated enzyme immobilization and modification of QDs.