Olympus
From Olympus
Olympus is an architecture for spoken dialog system created at Carnegie Mellon University. It is mainly designed to help researchers in conversational agents implement and test their ideas on full systems, without having to build them from scratch.
Contents |
Overview
The Olympus architecture (Bohus et al. 2007) incorporates a number of modules developed at CMU in past and current spoken dialogue systems research projects. Dialogue management is handled by RavenClaw (Bohus & Rudnicky 2009), a task-independent dialogue engine based on the AGENDA dialog manager (Wu & Rudnicky 1999) first introduced as part of the CMU Communicator system. Low-level interaction management (e.g. exact timing of start and end of utterances, handling of interruptions, etc) is performed by the Apollo interaction manager (Raux & Eskenazi 2007). For speech recognition, Olympus currently supports engines from the CMU Sphinx family (Sphinx 2, Sphinx 3, PocketSphinx), and more work is being done to incorporate other engines as well. Natural language understanding is done by Phoenix (Ward & Issar 1994), a robust parser based on CFG-like grammars. Natural language generation uses the Rosetta template-based generation system, while Kalliope, the synthesis interface currently allows the use of SAPI 5-compliant TTS engines, Festival and the proprietary engine Cepstral Swift. The communication between the different modules is handled by the MIT/MITRE Galaxy Communicator architecture.
Funding for the development of Olympus and the systems that is based on has been provided by a number of sponsors, including the Defense Advanced Research Projects Agency for CMU Communicator and CALO, the Office of Naval Research for LARRI, and from the National Science Foundation for the Let's Go project (grants number 0208835 and 0741773). The Boeing Corporation has in part supported the development of Olympus through the Treasure Hunt project.
Download
The Olympus system and some example applications are available under an open source license. Olympus may be accessed via a subversion repository.
Documentation
We are currently working on writing serious documentation for Olympus. Some of it is already (at least partially) available such as a tutorial and some reference pages. We will post announcements here and on the distribution mailing list as more of it gets completed.
Questions and Problems
For any additional information, as well as specific questions, send email to the developers mailing list (olympus-developers@@cs.cmu.edu).
Authors
Many many people have contributed to Olympus over the years (including prior to its existence...). The main contributors are:
Other (sometimes significant) contributions have been made by:
- Alan W Black
- Maxine Eskenazi
- Andrew Hoskins
- Ben Frisch
- Scott Judy
- Brian Langner
- Matthew Marge
- Udhyakumar Nallasamy
- Alexander I. Rudnicky
- Jahanzeb Sherwani
- Svetlana Stenchikova
- Yitao Sun
Systems
| Name | Description | Developers | Phone # |
|---|---|---|---|
| RoomLine | Provides support for conference room scheduling and reservation within the School of Computer Science at Carnegie Mellon University | Dan Bohus | N/A |
| Let's Go! Bus Information System | Provides access to bus route information in the greater Pittsburgh area | Antoine Raux, Dan Bohus, Brian Langner, Alan Black, Maxine Eskenazi | 412-268-3526 |
| TeamTalk | Spoken language command-and-control interface to a team of robots | June Sison, Satanjeev Banerjee, Thomas Harris, Matthew Marge, S.P.Kishore, Kerry Bodine, Dan Bohus | N/A |
| ConQuest | Telephone-based system that provides conference schedule information (deployed during Interspeech 2006 and IJCAI 2007) | Rohit Kumar, Dan Bohus, Gopala Krishna, Venkatesh Keri, Sergio Grau Puerto, Stefanie Tomko, Antoine Raux | N/A |
| MeetingLine | Telephone-based system that provides access to recorded meeting information. | Dan Bohus, Satanjeev Banerjee | N/A |
| Intelligent Procedure Assistant | Multi-modal spoken dialogue system aimed at providing guidance and support to the astronauts on the ISS during the execution of procedural tasks and checklists | RIALIST group/NASA Ames, Greg Aist, Dan Bohus | N/A |
| LARRI | Multi-modal system that provides assistance to F18 aircraft personnel during the execution of maintenance tasks | Dan Bohus, Yitao Sun, Kayur Patel, Ananlada Chotimongkol | N/A |
| Sublime | PDA-based personal information management system. | Jahanzeb Sherwani | N/A |
| Vera | Telephone-based taskable agent that can be instructed to deliver messages to a third party and make wake-up calls. | Ulas Bardak, Scott Judy, Vasco Pedro, Tal Blum, Jeongwoo Ko, Ryosuke Miyata | N/A |
| Madeleine | Written language (text I/O) spoken dialog system for medical diagnosis (toy domain) | Dan Bohus | N/A |
Development
References
Bohus, Dan & Alexander I. Rudnicky (2009), "The RavenClaw dialog management framework: Architecture and systems", Computer Speech & Language
Bohus, Dan & Alexander I. Rudnicky (2003), "RavenClaw: Dialog Management Using Hierarchical Task Decomposition and an Expectation Agenda", Eurospeech 2003
Bohus, Dan; Antoine Raux; Thomas K. Harris; Maxine Eskenazi & Alexander I. Rudnicky (2007), "Olympus: an open-source framework for conversational spoken language interface research", Bridging the Gap: Academic and Industrial Research in Dialog Technology workshop at HLT/NAACL 2007
Raux, Antoine & Maxine Eskenazi (2007), "A Multi-Layer Architecture for Semi-Synchronous Event-Driven Dialogue Management", IEEE Automatic Speech Recognition and Understanding Workshop
Rudnicky & Wei Xu (December 1999, p. I-337), "An agenda-based dialog management architecture for spoken language systems", IEEE ASRU Workshop
Ward, Wayne & Sunil Issar (1994), "Recent improvements in the CMU spoken language understanding system", ARPA Human Language Technology Workshop
