The prosody module

Abstract. We describe the acoustic-prosodic and syntactic-prosodic annotation and classi-fication of boundaries, accents and sentence mood integrated in the Verbmobil system for the three languages German, English, and Japanese. For the acoustic-prosodic classification, a large feature vector with normalized prosodic features is used. For the three languages, a multilingual prosody module was developed that reduces memory requirement considerably, compared to three monolingual modules. For classification, neural networks and statistic lan-guage models are used.