 |
Projects
|
Since 1994 I have been working in the following projects:
|
 |
Natural Language Generation for Acoustic Output (1994-95) laid the foundations for a system for the concept-to-speech system VieCtoS that
combines natural language generation with speech synthesis. The project focussed on architectural
issues, especially on laying the foundations for a representation that combines syntactic and semantic
structures allowing for the derivation of prosodic features relevant for better intonation.
|
|
|
VieCtoS (Vienna Concept-to-Speech System) (1996-98) is a
Concept-to-Speech system focussing on how to produce correct
prosody and intonation. The area of application is the automatic
production of weather forecasts from formatted data.
|
|
|
SpeeDurCont (1999-2001)
This project deals with the investigation of segmental duration in Austrian German. From a theoretical point of view it aims for a better understanding of the factors that influence duration as well as their interaction. From a practical point of view the results are to be used in order to improve the prosodic quality of our synthesizer.
|
|
|
NECA (A Net Environment for Embodied Emotional Conversational Agents) (2001-2004)
The aim of this EU-IST project is the development of a novel platform for the implementation of avatars. The modelling of emotions and their expression via affective speech synthesis, mimic, and gesture are central tasks within this project.
|
|
|
HUMAINE (2005-2007) This EU 6th Framework "Network of Excellence" aims to lay the foundations for the development of systems that can register, model and/or influence human emotional and emotion-related states and processes - 'emotion-oriented systems'.
|
|
|
SEMPRE (Semantically Aware Profiling for Recommenders) (2008-2009) This project is concerned with the exploitation of information gathered from the Internet for recommender systems. My main concern within this project is the application of opinion mining in the music-domain.
|
|
|
CyberEmotions (Collective Emotions in Cyberspace) (2009-2011) The project investigates the role of collective emotions in creating, forming and breaking-up of e-communities. In this fremework I am concerned with the task of detecting and classifying emotions in texts (postings in online discussions).
|
|
Research
Interests
|
Issues I have been working on:
|
|
|
Computational Morphology: I was especially concerned with finite-state-morphology.
In addition my MA-thesis dealt with issues of morphology in HPSG. |
|
|
Speech synthesis and more specifically Concept to Speech Synthesis: I mainly worked on the modelling of prosody, both in terms of intonation and duration |
|
|
I performed some "shallow research" in human factors and
speech recognition, synthesis of affective speech, and Wizard-of-Oz
simulations of Spoken Dialogue Systems. |
|
|
Embodied Conversational Agents ("avatars") and talking heads: focusing on the automatic generation of speech-accompanying gestures, facial expressions and postures. |
|
|
Opinion mining in reviews of movies and music albums. |
|
|
Emotion classification in texts. In the context of the CyberEmotions project I am currently working on the task of classifying texts, typically user postings in online fora and newsgroups, for their positive or negative valence. |
