Search

Your search keyword '"Schwenk, Holger"' showing total 175 results
Start Over Author "Schwenk, Holger"
175 results on '"Schwenk, Holger"'

161. Seven Doubly Bridged Ferrocene Units in a Cycle

Author

Grossmann, Birgit, primary, Heinze, Jürgen, additional, Herdtweck, Eberhardt, additional, Köhler, Frank H., additional, Nöth, Heinrich, additional, Schwenk, Holger, additional, Spiegler, Michael, additional, Wachter, Wolfgang, additional, and Weber, Bernd, additional

Published
1997
Full Text
View/download PDF

167. Advances in Transcription of Broadcast News and Conversational Telephone Speech Within the Combined EARS BBN/LIMSI System.

Author

Matsoukas, Spyros, Gauvain, Jean-Luc, Adda, Gilles, Colthurst, Thomas, Chia-Lin Kao, Kimball, Owen, Lamel, Lori, Lefevre, Fabrice, Ma, Jeff Z., Makhoul, John, Long Nguyen, Prasad, Rohit, Schwartz, Richard, Schwenk, Holger, and Bing Xiang

Subjects
TRANSCRIPTION (Linguistics), COMPUTER-aided transcription systems, BROADCAST journalism, BROADCASTING industry, MARKOV processes
Abstract

This paper describes the progress made in the transcription of broadcast news (BN) and conversational telephone speech (CTS) within the combined BBN/LIMSI system from May 21102 to September 2004. During that period, BBN and LIMSI collaborated in an effort to produce significant reductions in the word error rate (WER), as directed by the aggressive goals of the Effective, Affordable. Reusable, Speech-to-text [Defense Advanced Research Projects Agency (DARPA EARS) program. The paper focuses on general modeling techniques that led to recognition accuracy improvements, as well as engineering approaches that enabled efficient use of large amounts of training data and fast decoding architectures. Special attention is given on efforts to integrate components of the BBN and LIMSI systems, discussing the tradeoff between speed and accuracy for various system combination strategies. Results on the EARS progress test sets show that the combined BBN/LIMSI system achieved relative reductions 0f47% and 51% on the BN and CTS domains, respectively. [ABSTRACT FROM AUTHOR]

Published
2006
Full Text
View/download PDF

171. Introduction to the special issue on deep learning approaches for machine translation

Author

Ruiz Costa-Jussà, Marta, Allauzen, Alexandre, Barrault, loïc, Cho, Kyunghun, Schwenk, Holger, Universitat Politècnica de Catalunya. Departament de Teoria del Senyal i Comunicacions, and Universitat Politècnica de Catalunya. VEU - Grup de Tractament de la Parla

Subjects
Traducció automàtica, Aprenentatge automàtic, Informàtica::Intel·ligència artificial [Àrees temàtiques de la UPC], Deep learning, Neural Networks (Computer), Machine translation, Machine translating
Abstract

Deep learning is revolutionizing speech and natural language technologies since it is offering an effective way to train systems and obtaining significant improvements. The main advantage of deep learning is that, by developing the right architecture, the system automatically learns features from data without the need of explicitly designing them. This machine learning perspective is conceptually changing how speech and natural language technologies are addressed. In the case of Machine Translation (MT), deep learning was first introduced in standard statistical systems. By now, end-to-end neural MT systems have reached competitive results. This special issue introductory paper addresses how deep learning has been gradually introduced in MT. This introduction covers all topics contained in the papers included in this special issue, which basically are: integration of deep learning in statistical MT; development of the end-to-end neural MT system; and introduction of deep learning in interactive MT and MT evaluation. Finally, this introduction sketches some research directions that MT is taking guided by deep learning.

172. Synthesis and structures of bis(tetramethylpiperidino)aluminum halides. X-ray crystal structures of tmp2AlX (X = Cl, Br, I) and [tmp2Al(m-F)]2

Author

Krossing, Ingo, Noth, Heinrich, Tacke, Christiane, Schmidt, Martin, and Schwenk, Holger

Abstract

N-lithio-2,2,6,6-tetramethylpiperidine [Li(tmp)] reacts with AlX3 (X = Cl, Br) in Et2O/n-hexane soln. to generate of substitution and ether cleavage products, [(tmp)Al(X)(m-OEt)]2 (X = Cl, Br). However, when the reaction is allowed to proceed in n-hexane alone, an almost quant. yield of (tmp)2AlX (X = Cl, Br, I) is obtained. According to 27Al NMR, mass spectroscopy, cryoscopy, and x-ray crystal structure detns., these compds. are monomeric in the solid state, in soln., and in gas phase. (Tmp)2AlBr reacts with AgBF4 yielding the fluoride-bridged dimer [(tmp)2AlF]2 as shown by x-ray crystal structure detn. [on SciFinder (R)]

173. Sieben doppelt verbrückte Ferroceneinheiten in einem Ring

Author

Grossmann, Birgit, Heinze, Jürgen, Herdtweck, Eberhardt, Köhler, Frank H., Nöth, Heinrich, Schwenk, Holger, Spiegler, Michael, Wachter, Wolfgang, and Weber, Bernd

Abstract

Was hier wie ein Schaufelrad aussieht, sind sieben Ferroceneinheiten, deren jede durch vier Me2Si‐Gruppen mit zwei benachbarten verknüpft ist. Der Ring, der in einer Polygemination — das ist die Reaktion eines Übergangsmetallhalogenids mit einem Biscyclopentadienyl‐Dianion zu einem Polymer — gebildet wird, hat (idealisiert) die seltene C7h‐Symmetrie. Die Verbindung ist bis 360°C stabil und kann in drei Schritten bis zum Heptakation oxidiert werden.

Published
1997
Full Text
View/download PDF

174. Tris(tri-tert-butylsilyl)digallanyl (tBu3Si)3Ga2: A New Type of Compound for a Heavy Group 13 Element

Author

Wiberg, Nils, Amelunxen, Kerstin, Nöth, Heinrich, Schwenk, Holger, Kaim, Wolfgang, Klein, Axel, and Scheiring, Thomas

Abstract

Blue-black trisupersilyldigallanyl, a new metastable inorganic radical, forms in the reaction of GaCl3with tBu3SiNa in pentane at room temperature according to Equation (a). The X-ray structure analysis reveals that it contains a planar Si2GaIIGaIand a nearly linear GaIIGaISi framework; according to the ESR spectrum the unpaired electron resides in both Ga atoms. Surprisingly, the analogous reaction of AlCl3and tBu3SiNa (Al instead of Ga in Equation (a)) does not lead to trisupersilyldialanyl, but to red tetrasupersilyldialane (tBu3Si)2Al–Al(SitBu3)2.

Published
1997
Full Text
View/download PDF

175. Beyond English-Centric Multilingual Machine Translation.

Author

Fan, Angela, Bhosale, Shruti, Schwenk, Holger, Zhiyi Ma, El-Kishky, Ahmed, Goyal, Siddharth, Baines, Mandeep, Celebi, Onur, Wenzek, Guillaume, Chaudhary, Vishrav, Goyal, Naman, Birch, Tom, Liptchinsky, Vitaliy, Edunov, Sergey, Grave, Edouard, Auliy, Michael, and Joulin, Armand

Subjects
*MACHINE translating
Abstract

Existing work in translation demonstrated the potential of massively multilingual machine translation by training a single model able to translate between any pair of languages. However, much of this work is English-Centric, training only on data which was translated from or to English. While this is supported by large sources of training data, it does not reflect translation needs worldwide. In this work, we create a true Many-to-Many multilingual translation model that can translate directly between any pair of 100 languages. We build and open-source a training data set that covers thousands of language directions with parallel data, created through large-scale mining. Then, we explore how to effectively increase model capacity through a combination of dense scaling and language-specific sparse parameters to create high quality models. Our focus on non-English-Centric models brings gains of more than 10 BLEU when directly translating between non-English directions while performing competitively to the best single systems from the Workshop on Machine Translation (WMT). We open-source our scripts so that others may reproduce the data, evaluation, and final m2m-100 model: https://github.com/pytorch/fairseq/tree/master/examples/m2m_100. [ABSTRACT FROM AUTHOR]

Published
2021

Catalog

Books, media, physical & digital resources
See catalog results