Interspeech'2005 - Eurospeech

Lisbon, Portugal
September 4-8, 2005

Tree-Based Prediction of Prosodic Phrase Breaks on Top of Shallow Textual Features

Gerasimos Xydas (1), Panagiotis Zervas (2), Georgios Kouroupetroglou (1), Nikolaos Fakotakis (2), George Kokkinakis (2)

(1) University of Athens, Greece; (2) University of Patras, Greece

This paper reports on the evaluation of automatic prosodic phrase break assignment. We utilize two tree-structured predictors, the commonly used CART and a C4.5, to predict break placement from sequences of easily to extract shallow textual features. We are experimenting with two 500-utterance prosodic corpora developed by two Greek universities that originate from different domains in order to focus on the differences in prediction between generic and limited domain datasets. The evaluation shows that while the limited dataset achieves better accuracy than the generic one in the CART case, this difference is lowered with the introduction of C4.5. Minor breaks proved to be the most difficult class to predict in CART case, while we achieved a 50% improvement with C4.5.

Full Paper

Bibliographic reference.  Xydas, Gerasimos / Zervas, Panagiotis / Kouroupetroglou, Georgios / Fakotakis, Nikolaos / Kokkinakis, George (2005): "Tree-based prediction of prosodic phrase breaks on top of shallow textual features", In INTERSPEECH-2005, 3237-3240.