5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

On The Use of Prosody in a Speech-To-Speech Translator

Volker Strom (1), Anja Elsner (1), Wolfgang Hess (1), Walter Kasper (4), Alexandra Klein (2), Hans Ulrich Krieger (4), Jörg Spilker (3), Hans Weber (3), Gunther Gorz (3)

(1) Institute of Communications Research and Phonetics (IKP) University of Bonn, Germany (2) University of Wien, Austrian Research Institute of Artificial Intelligence, Wien, Austria (3) University of Erlangen-Nürnberg, Computer Science Institute (AI), Germany (4) German Research Center for AI, DFKI GmbH, Saarbrucken, Germany

In this paper a speech-to-speech translator from German to English is presented. Beside the traditional processing steps it takes advantage of acoustically detected prosodic phrase boundaries and focus. The prosodic phrase boundaries reduce search space during syntactic parsing and rule out analysis trees during semantic parsing. The prosodic focus faciliates a "shallow" translation based on the best word chain in cases where the deep analysis fails.

Full Paper

Bibliographic reference.  Strom, Volker / Elsner, Anja / Hess, Wolfgang / Kasper, Walter / Klein, Alexandra / Krieger, Hans Ulrich / Spilker, Jörg / Weber, Hans / Gorz, Gunther (1997): "On the use of prosody in a speech-to-speech translator", In EUROSPEECH-1997, 1479-1482.