|
Home >
Research >
Techonology and Communication >
Computer Science, Technology and Multimedia >
UOC conference proceedings >
Second International Workshop on Free/Open-SourceRule-Based Machine Translation, 2011 >
Please use this identifier to cite or link to this item:
http://hdl.handle.net/10609/5647
|
| Title: | Bootstrapping a statistical speech translator from a rule-based one |
| Autor: | Rayner, Manny Estrella, Paula Bouillon, Pierrette |
| Otros: | International Workshop on Free/Open-Source Rule-Based Machine Translation (2nd : 2011 : Barcelona) |
| Keywords: | Statistical bootstrapping |
| Issue Date: | 20-Jan-2011 |
| Editorial: | Universitat Oberta de Catalunya |
| Type: | Conferencia |
| Citación: | Rayner, Manny; Estrella, Paula; Bouillon, Pierrette (2011, January). "Bootstrapping A Statistical Speech Translator From A Rule-Based One". Proceedings of the Second International Workshop on Free/Open-Source Rule-Based Machine Translation (2011: Barcelona). <http://hdl.handle.net/10609/5647> |
| Abstract: | We describe a series of experiments in which we start with English to French and English to Japanese versions of an Open Source rule-based speech translation system for a medical domain, and bootstrap correspondign statistical systems. Comparative evaluation reveals that the rule-based systems are still significantly better than the statistical ones, despite the fact that considerable effort has been invested in tuning both the recognition and translation components; also, a hybrid system only marginally improved recall at the cost of a los in precision. The result suggests that rule-based architectures may still be preferable to statistical ones for safety-critical speech translation tasks. |
| Language: | eng |
| URI: | http://hdl.handle.net/10609/5647 |
| ISBN: | 978-84-693-9517-2 |
| Appears in Collections: | Second International Workshop on Free/Open-SourceRule-Based Machine Translation, 2011
|
Author names in FriendFeed
This item is licensed under a Creative Commons License
Items in Repository e are protected by copyright, with all rights reserved, unless otherwise indicated.
|