Project C01

VisAnnotate

Principal Investigators: Miriam Butt
Daniel Keim
Research Staff: Tina Bögel
Dominik Sacha

Project Description:
The aim of the project is to develop a system that meets the needs of the projects A01, A02 and B01. In the project, information on different aspects of the speech signal (voice quality, length, intensity, fundamental frequency and pauses) will be analysed and interpreted from a semantic-pragmatic view point in order to accomplish a reasonable coverage of bigger amounts of spoken data and its different aspects. The focus here lies on analysing and extending existing system based on the needed linguistic categories in order to, on the basis of extracted data, make predictions for linguistic analyses (semantics, pragmatics, syntax) and based on this to develop an annotational system in cooperation with A01 and B01. A visual data analysis system will be developed for the annotation component enabling the user to automatically make linguistic predictions and to interactively correct them. Through automatic learning, the system can thus be trained and refined step by step. Furthermore, a visual representation of the data will be developed and implemented to allow for a quick classification of the analysed statement within discourse analysis.