Sangyun Hahn

Learn More
RASR is the open source version of the well-proven speech recognition toolkit developed and used at RWTH Aachen University. The current version of the package includes state of the art speech recognition technology for acoustic model training and decoding. Speaker adaptation, speaker adaptive training, unsupervised training, discriminative training, lattice(More)
F. Abe, H. Akimoto, A. Akopian, M. G. Albrow, A. Amadon, S. R. Amendolia, D. Amidei, J. Antos, S. Aota, G. Apollinari, T. Arisawa, T. Asakawa, W. Ashmanskas, M. Atac, P. Azzi-Bacchetta, N. Bacchetta, S. Bagdasarov, M. W. Bailey, P. de Barbaro, A. Barbaro-Galtieri, V. E. Barnes, B. A. Barnett, M. Barone, G. Bauer, T. Baumann, F. Bedeschi, S. Behrends, S.(More)
Access to graphical images (bar charts, diagrams, line graphs, etc.) that are in a tactile form (representation through which content can be accessed by touch) is inadequate for students who are blind and take mathematics, science, and engineering courses. We describe our analysis of the current work practices of tactile graphics specialists who create(More)
We present a measurement of the W-boson mass using 200 pb{-1} of data collected in pp[over ] collisions at sqrt[s]=1.96 TeV by the CDF II detector at run II of the Fermilab Tevatron. With a sample of 63 964 W-->enu candidates and 51 128 W-->munu candidates, we measure M_{W}=80 413+/-34{stat}+/-34{syst}=80,413+/-48 MeV/c;{2}. This is the most precise single(More)
We report on ν(e) and ν(e) appearance in ν(μ) and ν(μ) beams using the full MINOS data sample. The comparison of these ν(e) and ν(e) appearance data at a 735 km baseline with θ13 measurements by reactor experiments probes δ, the θ23 octant degeneracy, and the mass hierarchy. This analysis is the first use of this technique and includes the first accelerator(More)
We present the first search for heavy, long-lived particles that decay to photons at a hadron collider. We use a sample of gamma + jet + missing transverse energy events in pp[over] collisions at square root[s] = 1.96 TeV taken with the CDF II detector. Candidate events are selected based on the arrival time of the photon at the detector. Using an(More)
We present measurements of the forward-backward charge asymmetry in top pair production using 1.9 fb;{-1} of pp[over ] collisions at sqrt[s]=1.96 TeV recorded with the Collider Detector at Fermilab II. Correcting for acceptance and measurement dilutions we obtain parton-level asymmetries of A_{FB};{pp[over ]}=0.17+/-0.08 in the pp[over ] frame and(More)
We present a precision measurement of the top-quark mass using the full sample of Tevatron √s = 1.96 TeV proton-antiproton collisions collected by the CDF II detector, corresponding to an integrated luminosity of 8.7 fb(-1). Using a sample of tt¯ candidate events decaying into the lepton+jets channel, we obtain distributions of the top-quark masses and the(More)
Several semi-supervised learning methods have been proposed to leverage unlabeled data, but imbalanced class distributions in the data set can hurt the performance of most algorithms. In this paper, we adapt the new approach of contrast classifiers for semi-supervised learning. This enables us to exploit large amounts of unlabeled data with a skewed(More)
We report measurements of oscillation parameters from ν(μ) and ν(μ) disappearance using beam and atmospheric data from MINOS. The data comprise exposures of 10.71×10(20) protons on target in the ν(μ)-dominated beam, 3.36×10(20) protons on target in the ν(μ)-enhanced beam, and 37.88 kton yr of atmospheric neutrinos. Assuming identical ν and ν oscillation(More)