- Components of the Litebody system, and their features, include:
- Litebody Protocol: an http or https based client-server protocol for delivering audio and animation scripts from a server to an ECA client, and for delivering user input from client to server.
- Litebody Client: a Flash-based ECA client, which supports a core set of nonverbal behavior. Our current implementations�including the character animation distributed with Litebody�are as small as 500K, allowing reasonable download times even on slow network connections. It can allow multiple-choice or free-text user input, and has a plug-in architecture that can be used to provide additional input methods.
- Litebody Server Framework: used to build Java-based web applications which implement the Litebody Protocol. The framework abstracts low-level details of the protocol and is intended to allow the use of arbitrary dialogue engines with Litebody. We also provide an open-source dialogue engine, DTask, which is based on hierarchical task decomposition and a public standard task description language, and readily interoperates with Litebody.
- Litebody TTS Server: handles text-to-speech synthesis, and is responsible for generating audio files and timing information. We provide a Java-based implementation which currently can use either FreeTTS (open-source) or Loquendo (with the addition of licensed software), with a Microsoft SAPI interface in development.
- Bickmore, T., Schulman, D. and Shaw, G. (2009) DTask & LiteBody: Open Source, Standards-based Tools for Building Web-deployed Embodied Conversational Agents Proceedings of Intelligent Virtual Agents, Amsterdam. PDF
LiteBody by Northeastern University Relational Agents Group is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License. Based on a work at litebodysuite.sourceforge.net
We also request that any publications covering work based on LiteBody cite the Bickmore, et al, 2009 IVA paper (above).