ETRW on Dialogue and Prosody

September 1-3, 1999
Veldhoven, The Netherlands

On the Use of Prosody in Automatic Dialogue Understanding

Elmar Nöth, Anton Batliner, Volker Warnke, J. Haas, M. Boros, Jan Buckow, Richard Huber, F. Gallwitz, Matthias Nutt, Heinrich Niemann

University of Erlangen-Nuremberg and FORWISS, Erlangen, Germany

In this paper, we show how prosodic information can be used in automatic dialogue systems and give some examples of promising new approaches. Most of these examples are taken from our own work in the VERBMOBIL speech-to-speech translation system and the EVAR train timetable dialogue system. In a ‘prosodic orbit’, we first present units, phenomena, annotations and statistical methods from the signal (acoustics) to the dialogue understanding phase. We show then, how prosody can be used together with other knowledge sources for the task of resegmentation and how an integrated approach leads to better results than a sequential use of the different knowledge sources; then we present a hybrid approach which is used to perform a shallow parsing and which uses prosody to guide the parsing; finally, we show how a critical system evaluation can help to improve the overall performance of automatic dialogue systems.

Full Paper (PDF)   Zipped Postscript

Bibliographic reference.  Nöth, Elmar / Batliner, Anton / Warnke, Volker / Haas, J. / Boros, M. / Buckow, Jan / Huber, Richard / Gallwitz, F. / Nutt, Matthias / Niemann, Heinrich (1999): "On the use of prosody in automatic dialogue understanding", In DIAPRO-1999, 25-34.