Interview with our partners

Jan. 4, 2017, noon

The social challenge of this project has an increasing importance: different research areas are integrated to improve a real system able to overcome linguistic barriers for migrants looking for health care assistance in Europe. Within the European funded project KRISTINA, technical partners are constantly working in many areas in order to make this conversation more natural, improving the expressions and the ability to recognize and interpret emotions and human feelings. This first prototype is used to demonstrate the current state-of-the-art in natural conversation and it can be
considered as a working basis on which enhancements will be build, each aimed at providing a more natural conversation. The agent will recognize emotions in the voice, face and body stance of the conversation partner. During the third plenary meeting, some members of our project, have been interviewed and they explained us the current state of their research, the main challenges and expectations.

Leo Wanner, main responsible and coordinator of the KRISTINA project, ensures that the general objective of the first prototype has been achieved: users can conduct a basic dialogue with the system and this interaction is now supported by an overall structure that includes the interaction and the exchange of information between each system component. The first prototype was not expected to support a flexible dialogue but partners are now on the right track and they will continue to work, improving the functionalities and the quality of each individual component. First of all it is necessary to improve the accuracy of the speech recognition, because the agent takes too long to analyse the sentences and it often misunderstand the user. Secondly KRISTINA should better process what the end-user is saying, understanding and providing the right information with an appropriate reaction. Finally, “flexibility” is the key to the future system development: the users should communicate with the virtual agent in a more flexible way and in their native language.

Watch the interview with our partners!

multilingual intelligent embodied agent social competence adaptive dialogue expressive speech recognition and synthesis discourse generation vocal facial and gestural social and emotional cues