Listening Human(oid)s: Analysis, Modeling and Synthesis of Non-verbal Listening Behavior

Description of research

In everyday conversation we signal understanding, agreement and affective responses to the speaker while listening. We nod at certain moments, we frown our eye brows or we smile. These signals play an important role in the flow of the conversation. To be able to have a smooth conversation with a virtual human, the system needs to be able to produce the same backchannel communication signals at the appropriate times.

Therefore, I aim to build a believable model of non-verbal listening behavior for virtual humans. The model will be based on a multimodal analysis of recorded human-to-human interactions. Methods that will be used during the project are literature study, multimodal corpus analysis, corpus based machine learning, artificial intelligence modeling and user evaluation of the model in an interactive system.

Advisor(s)

Prof.Dr.Ir. Anton Nijholt (Promotor)

Dr. Dirk Heylen (Daily Supervisor)

Duration

Aug 2009 – Aug 2013

Project

SEMAINE
SERA
SSPNET

Funding institution

EU FP7

Strategic Research Orientation

NICE

Links to relevant web pages:

http://wwwhome.cs.utwente.nl/~koki/

http://hmi.ewi.utwente.nl/person/Iwan%20de%20Kok

Pictures

	CTIT Home > Meet the PhDs > Iwan de Kok
Home CTIT symposium 2013 About CTIT News NIRICT Research Meet the PhDs Working at CTIT Internal Information Conferences Open Calls (updated May 3, 2013) Library and Videos Knowledge Transfer Links Sitemap Search	Listening Human(oid)s: Analysis, Modeling and Synthesis of Non-verbal Listening Behavior Description of research In everyday conversation we signal understanding, agreement and affective responses to the speaker while listening. We nod at certain moments, we frown our eye brows or we smile. These signals play an important role in the flow of the conversation. To be able to have a smooth conversation with a virtual human, the system needs to be able to produce the same backchannel communication signals at the appropriate times. Therefore, I aim to build a believable model of non-verbal listening behavior for virtual humans. The model will be based on a multimodal analysis of recorded human-to-human interactions. Methods that will be used during the project are literature study, multimodal corpus analysis, corpus based machine learning, artificial intelligence modeling and user evaluation of the model in an interactive system. Advisor(s) Prof.Dr.Ir. Anton Nijholt (Promotor) Dr. Dirk Heylen (Daily Supervisor) Duration Aug 2009 – Aug 2013 Project SEMAINE SERA SSPNET Funding institution EU FP7 Strategic Research Orientation NICE Links to relevant web pages: http://wwwhome.cs.utwente.nl/~koki/ http://hmi.ewi.utwente.nl/person/Iwan%20de%20Kok Pictures