System combination for machine translation for spoken and written language
| Autor: | Daniel Déchelotte, Hermann Ney, Evgeny Matusov, Holger Schwenk, Salim Roukos, Nicola Bertoldi, Gregor Leusch, José B. Mariño, M. Kolss, Young-Suk Lee, Marcello Federico, M. Paulik, Rafael E. Banchs |
|---|---|
| Přispěvatelé: | Universitat Politècnica de Catalunya. VEU - Grup de Tractament de la Parla, meignier, sylvain, Laboratoire d'Informatique de l'Université du Mans (LIUM), Le Mans Université (UM) |
| Jazyk: | angličtina |
| Rok vydání: | 2008 |
| Předmět: |
Natural-language
Acoustics and Ultrasonics Machine translation Computer science Speech recognition 02 engineering and technology computer.software_genre Speech-processing Machine translation software usability [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] Example-based machine translation Rule-based machine translation Speech translation Traducció automàtica 0202 electrical engineering electronic engineering information engineering Electrical and Electronic Engineering Language translation ComputingMilieux_MISCELLANEOUS business.industry 020206 networking & telecommunications Transfer-based machine translation [INFO.INFO-CL] Computer Science [cs]/Computation and Language [cs.CL] Computer-assisted translation Enginyeria de la telecomunicació::Processament del senyal::Processament de la parla i del senyal acústic [Àrees temàtiques de la UPC] 020201 artificial intelligence & image processing Artificial intelligence Machine-translation business computer Machine translating Natural language processing |
| Zdroj: | Recercat. Dipósit de la Recerca de Catalunya instname UPCommons. Portal del coneixement obert de la UPC Universitat Politècnica de Catalunya (UPC) IEEE Transactions on Audio, Speech and Language Processing IEEE Transactions on Audio, Speech and Language Processing, Institute of Electrical and Electronics Engineers, 2007, 16 (7), pp.1222--1237 |
| ISSN: | 1558-7916 |
| Popis: | This paper describes a recently developed method for computing a consensus translation from the outputs of multiple machine translation (MT) systems. A possibly new translation hypothesis can be produced as a result of this system combination algorithm. The consensus translation is computed by creating a confusion network and performing weighted majority voting, similarly to the well-established ROVER approach of (Fiscus 1997) for combining speech recognition hypotheses. To create the confusion network, pairwise word alignments of the original machine translation hypotheses are learned by using an enhanced statistical alignment algorithm that explicitly models word reordering. This is the first known application of this algorithm in the context of system combination. The context of a whole document of translations rather than a single sentence is taken into account to improve the alignment quality. The proposed alignment and voting approach was evaluated on several machine translation tasks, including a large vocabulary task. The method was also tested in the framework of multi- source and speech translation. Significant improvements in translation quality were achieved on all tasks. Here, we report experimental results for combining MT systems participating in the TC-STAR (speech translation) Project. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|