Agent Foundation Classes (AFC)
The AFC is a toolkit and library designed to make the construction of the many types of intelligent software agents a semi-automated process. One of the aims of the AFC is to extend agent technology use and construction to the broader public. The AFC toolkit comes with an instruction manual that programmers and naive users can use to build an easy-assembly system of agents that they can apply to their organizational needs.

Communicator
Explores advanced language-based interfaces, both speech-only and multi-modal. Our research includes dialog management, natural language generation and synthesis. We are interested in the integration of different levels of processing and in the detection and robust handling of misunderstandings between humans and machines.

Computer Music
This project is developing computer music and interactive performance technology to enhance human musical experience and creativity. Active areas include MUSART, an investigation of style-based music information retrieval, AuraRT, a distributed system for interactive real-time multimedia, and Audacity, an audio visualization and editing tool.

Jam-O-Drum
The Jam-O-Drum is a multi-player controller that encourages team building and collaboration. By combining novel input devices with real-time computer graphics on an integrated tabletop surface, the Jam-O-Drum allows people to explore interactive game play and collective music-making experiences. Players' actions, detected via embedded sensors, control game play, real-time computer graphics and musical interaction.

Matchmaker
The Matchmaker is an information agent that helps make connections between agents that request services and agents that provide services. The Matchmaker interacts with human requestors, serving as an electronic "yellow pages" of agent capabilities, matching service providers with service requestors based on agent capability descriptions.

Mobile Commerce Lab
The Mobile Commerce Lab explores the development of mobile Internet applications and services, placing a particular emphasis on the use of personalization, context-awareness and intelligent agents to overcome the limitations of mobile access devices.

Multidisciplinary University Research Initiative (MURI)
For this MURI project we have conducted multidisciplinary research aimed at incorporating intelligent agent assistants into human teams to increase the effectiveness of team decision-making in joint planning tasks.

MyCampus
MyCampus is a project that explores context-aware scenarios aimed at enhancing everyday campus life. The project combines a strong HCI perspective with the development of advanced Semantic Web services, intelligent agent functionality and learning technologies.

Non-Combatant Evacuation Operation (NEO)
The NEO project illustrates the use of agent technology for cooperatively planning and executing a hypothetical evacuation of US civilians from a Middle Eastern city in an escalating terrorism crisis. In NEO, numerous agents cooperate to effectively aid the human team, providing continuously updated information and responding to inputs from humans and their environment.

RETSINA
The Software Agents group at CMU's Robotics Institute envisions a world in which autonomous, intelligent software programs known as software agents undertake many operations performed by human users of the Web, as well many other tasks. The group has developed the RETSINA multi-agent system infrastructure and applied it and its agents to many domains.

Text Miner
Text Miner is a system that automatically classifies news and assigns values to articles. In the domain of stock portfolio management (http://www-2.cs.cmu.edu/~softagents/warren.html), software agents assist their human users to evaluate the risks associated with the individual companies in a portfolio, by reading, classifying and weighing electronic news articles, giving investors an indication of the financial outlook of companies.

Universal Speech Interface
The USI project is an attempt to design a universal speech interface allowing humans to communicate effectively, efficiently and effortlessly with simple machines, information services, and internet speech portals. In essence, we are trying to do for speech communication what Graffiti has done for mobile text entry and what the Xerox/Macintosh universal look-and-feel has done for GUIs.

WARREN
WARREN is an application of the RETSINA (http://www-2.cs.cmu.edu/~softagents/retsina.html) multi-agent architecture to the problem of financial portfolio management. The WARREN system deploys a number of different, autonomous software agents who aid human users to manage their stock portfolios.