is a real-time three dimensional
embodied conversational agent with a 3D model of a woman compliant with MPEG-4 animation standard. She is able to
communicate using a rich palette of verbal and nonverbal behaviours. Greta
can talk and simultaneously show facial expressions,
gestures, gaze, and head movements.
standard XML languages FML and BML
allows the user to define her communicative intentions and
Greta can be used with different external TTS softwares. Currently she can speak various languages: English, Italian,
French, German, Swedish and Polish.
Greta is used in various European projects: CALLAS, SEMAINE, HUMAINE, and national French projects: ISCC Apogeste, ANR MyBlog3D, ANR IMMEMO .
To download Greta please read here
Here, you will find Greta's online documentation
Visit the Greta's forum. Write any comments and questions here.
Who works on Greta:
Quoc Anh Le
Etienne de Sevin
- Papers on the GRETA Architecture:
M. Mancini, C. Pelachaud :
Dynamic behavior qualifiers for conversational agents.
Intelligent Virtual Agents, IVA'07, Paris, September 2007.
E. Bevacqua, M. Mancini, R. Niewiadomski, C. Pelachaud:
An expressive ECA showing complex emotions.
AISB'07 Annual convention, workshop ``Language, Speech and Gesture for Expressive Characters'', Newcastle UK, pp. 208-216, April 2007 (pdf)
E. Bevacqua, M. Mancini, C. Pelachaud:
Speaking with Emotions
AISB 2004 Convention: Motion, Emotion and Cognition, University of
Leeds, UK, 2004 (pdf)
B. Hartmann, M. Mancini, C. Pelachaud:
Formational Parameters and Adaptive Prototype Instantiation for MPEG-4
Compliant Gesture Synthesis.
Proceedings of Computer Animation 2002, IEEE Computer Society Press, 2002 (pdf)
Other files: click here