Research

About

I am a Research Associate at the Spoken Dialogue Systems Group at the University of Cambridge working with Prof. Steve Young. I have received my Diploma (MSc) in Computer Science from the Karlsruhe Institute of Technology (Germany) in 2010 and my doctorate in engineering (PhD) at the Dialogue Systems Group at Ulm University (Germany) in 2015. The topic of my dissertation was „User-centred Adaptive Spoken Dialogue Modelling“.

The general goal of my research is to create a system which allows humans to talk to machines in the same natural way as to humans. To reach this goal, it is essential to answer the following questions: which properties and abilities must a system have to act in a natural manner, which factors make the behaviour of the system to be perceived as natural, and how may this natural interaction be realized from a technical point of view. In other words, we have to understand what exactly the term “natural system behaviour” means and we have to identify and realise the technical requirements and find suitable dialogue models to represent all necessary aspects of the dialogues.

My current research focuses on open domain statistical spoken dialogue systems within the EPSRC project “Open Domain Statistical Spoken Dialogue Systems”. I am especially interested in modelling the reward signal for reinforcement learning settings to take account for effects which may arise due to the added complexity of multiple domains. Here, I believe that the solution is to model the problem in a more user-centred way.

Additionally, I am co-supervising four doctoral students of the Dialogue Systems Group at Ulm University and I am an external advisor for the research project “KRISTINA” which is funded by the H2020 programme of the European Union.


Professional Activities