Search

Applications which use human speech as an input require a speech interface with high recognition accuracy. The words or phrases in the recognised text are annotated with a machine-understandable meaning and linked to knowledge graphs for further processing by the target application. These semantic annotations of recognised words can be represented as a subject-predicate-object triples which collectively form a graph often referred to as a knowledge graph. This type of knowledge representation facilitates to use speech interfaces with any spoken input application, since the information is represented in logical, semantic form, retrieving and storing can be followed using any web standard query languages. In this work, we develop a methodology for linking speech input to knowledge graphs and study the impact of recognition errors in the overall process. We show that for a corpus with lower WER, the annotation and linking of entities to the DBpedia knowledge graph is considerable. DBpedia Spotlight, a tool to interlink text documents with the linked open data is used to link the speech recognition output to the DBpedia knowledge graph. Such a knowledge-based speech recognition interface is useful for applications such as question answering or spoken dialog systems., Comment: Under Review in International Workshop on Grounding Language Understanding, Satellite of Interspeech 2017

With the evolution of neural network based methods, automatic speech recognition (ASR) field has been advanced to a level where building an application with speech interface is a reality. In spite of these advances, building a real-time speech recogniser faces several problems such as low recognition accuracy, domain constraint, and out-of-vocabulary words. The low recognition accuracy problem is addressed by improving the acoustic model, language model, decoder and by rescoring the N-best list at the output of the decoder. We are considering the N-best list rescoring approach to improve the recognition accuracy. Most of the methods in the literature use the grammatical, lexical, syntactic and semantic connection between the words in a recognised sentence as a feature to rescore. In this paper, we have tried to see the semantic relatedness between the words in a sentence to rescore the N-best list. Semantic relatedness is computed using TransE~\cite{bordes2013translating}, a method for low dimensional embedding of a triple in a knowledge graph. The novelty of the paper is the application of semantic web to automatic speech recognition.

Recently, there has been a growing interest in the study of deformed uranium-uranium ($U$-$U$) collisions in its various geometrical configurations due to their usefulness in understanding the different aspects of quantum chromodynamics (QCD). In this paper we have studied the particle production in deformed $U$-$U$ collisions at $\sqrt{s_{NN}}$ = $193$ GeV using modified wounded quark model (WQM). At first, we have shown the variation of quark-nucleus inelastic scattering cross-section ($\sigma_{qA}^{in}$) with respect to centralities for various geometrical orientations of $U$-$U$ collisions in WQM. After that we have calculated the pseudorapidity density ($dn_{ch}/d\eta$) within WQM using two-component prescription. Further we have calculated the transverse energy density distribution ($dE_{T}/d\eta$) along with the ratio of transverse energy to charged hadron multiplicity ($E_{T}/N_{ch}$) for $U$-$U$ collisions and compared them with the corresponding experimental data. We have shown the scaling behavior of $dn_{ch}/d\eta$ and $dE_{T}/d\eta$ for different initial geometry of $U$-$U$ collision with respect to $p$-$p$ data at $\sqrt{s_{NN}}=200$ GeV. Furthermore we have shown the Bjorken energy density achieved in $U$-$U$ collisions for various configurations and compared them with experimental data of $Au$-$Au$ at 200 GeV. We observe that the present model suitably describes the experimental data for minimum bias geometrical configuration of $U$-$U$ collisions. An estimate for various observables in different initial geometries of $U$-$U$ collisions is also presented which will be tested in future by experimental data., Comment: 23 pages, 12 figures, Accepted in EPJ Plus

Substantial experimental and theoretical efforts worldwide are devoted to explore the phase diagram of strongly interacting matter. At LHC and top RHIC energies, QCD matter is studied at very high temperatures and nearly vanishing net-baryon densities. There is evidence that a Quark-Gluon-Plasma (QGP) was created at experiments at RHIC and LHC. The transition from the QGP back to the hadron gas is found to be a smooth cross over. For larger net-baryon densities and lower temperatures, it is expected that the QCD phase diagram exhibits a rich structure, such as a first-order phase transition between hadronic and partonic matter which terminates in a critical point, or exotic phases like quarkyonic matter. The discovery of these landmarks would be a breakthrough in our understanding of the strong interaction and is therefore in the focus of various high-energy heavy-ion research programs. The Compressed Baryonic Matter (CBM) experiment at FAIR will play a unique role in the exploration of the QCD phase diagram in the region of high net-baryon densities, because it is designed to run at unprecedented interaction rates. High-rate operation is the key prerequisite for high-precision measurements of multi-differential observables and of rare diagnostic probes which are sensitive to the dense phase of the nuclear fireball. The goal of the CBM experiment at SIS100 (sqrt(s_NN) = 2.7 - 4.9 GeV) is to discover fundamental properties of QCD matter: the phase structure at large baryon-chemical potentials (mu_B > 500 MeV), effects of chiral symmetry, and the equation-of-state at high density as it is expected to occur in the core of neutron stars. In this article, we review the motivation for and the physics programme of CBM, including activities before the start of data taking in 2022, in the context of the worldwide efforts to explore high-density QCD matter., Comment: 15 pages, 11 figures. Published in European Physical Journal A

The charged particle multiplicity ($n_{ch}$) and pseudorapidity density $(dn_{ch}/d\eta)$ are key observables to characterize the properties of matter created in heavy ion collisions. The dependence of these observables on collision energy and the collision geometry are a key tool to understand the underlying particle production mechanism. Recently a lot of focus has been made on asymmetric and deformed nuclei collisions since these collisions can provide a deeper understanding about the nature of quantum chromodynamics (QCD). On phenomenological perspective a unified model which describes the experimental data coming from various kind of collision experiments, is much needed to provide the physical insights about the production mechanism. In this paper, we have calculated the charged hadron multiplicities for nucleon-nucleus (such as proton-lead $(p-Pb)$ and asymmetric nuclei collisions like deutron-gold $(d-Au)$, and copper-gold $(Cu-Au)$ within a new version of wounded quark model (WQM) and shown their variation with respect to centrality. Further we have used a suitable density function within our WQM to calculate pseudorapidity density of charged hadrons at mid-rapidity in the collisions of deformed uranium nuclei. We found that our model with suitable density functions describes the experimental data for symmetric, asymmetric and deformed nuclei collisions simultaneously over a wide range of collision energy., Comment: 25 pages, 15 figures, 3 tables

Question Answering (QA) systems are becoming the inspiring model for the future of search engines. While recently, underlying datasets for QA systems have been promoted from unstructured datasets to structured datasets with highly semantic-enriched metadata, but still question answering systems involve serious challenges which cause to be far beyond desired expectations. In this paper, we raise the challenges for building a Question Answering (QA) system especially with the focus of employing structured data (i.e. knowledge graph). This paper provide an exhaustive insight of the known challenges, so far. Thus, it helps researchers to easily spot open rooms for the future research agenda., Comment: Submitted to Question Answering And Activity Analysis in Participatory Sites (Q4APS) 2016

An extensive analysis of individual high multiplicity events produced in 158 A GeV /c 208Pb- 208Pb collisions is carried by adopting different methods to examine the anomalous behavior of these rare events. A method of selecting the events with densely populated narrow regions or spikes out of a given sample of collision events is discussed.Employing this approach two events with large spikes in their eta- and phi- distributions are selected for further analysis. For the sake of comparison, another two events which do not exhibit such spikes are simultaneously analyzed. The findings suggest that the systematic studies of particle density fluctuations in one- and two-dimensional phase-spaces and comparison with those obtained from the studies of correlation free Monte Carlo events, would be useful for identifying the events with large dynamical fluctuations. Formation of clusters or jet like phenomena in multihadronic final states in individual events is also discussed and the experimental findings are compared with the independent particle emission hypothesis by carrying out Monte Carlo simulations.