Janus - towards multilingual spoken language translation
Bernhard Suhm, Petra Geutner, Thomas Kemp, Ivica Rogina, Tanja Schultz, Sloboda, Tilo, Monika Woszczyna, Waibel, Alexander, T. Sloboda · Repository KITopen (Karlsruhe Institute of Technology) · 1995
In our effort to build spoken language translation systems we have extended our JANUS system to process spontaneoushuman--human dialogs in a new domain, two people trying to schedule a meeting. Trained on an initial database JANUS-2 is able to translate English and German spoken input in either English, German, Spanish, Japanese or Korean output. To tackle the difficulty of spontaneous human--human dialogs we improved the JANUS-2 recognizer along its three knowledgesourcesacousticmodels, dictionary andlanguage models. We developed a robust translation system which performs semantic rather than syntactic analysis and thus is particulary suited to processing spontaneous speech. We describe repair methods to recover from recognition errors. 1. Introduction JANUS [1, 2] has been among the first systems attempting to provide spoken language translation. While the previous JANUS-1 system processed syntactically wellformed read speech over a 500 word vocabulary, JANUS-2 operates on spontaneo...