Recently I put this one together:
- Olympus
CMU LTI, Dan Bohus (RavenClaw)
Olympus is an architecture for spoken dialog system created at Carnegie Mellon University (CMU). 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.
The Olympus architecture (Bohus et al. 2007) incorporates a number of modules developed at CMU in past and current spoken dialogue systems research projects. - Galatea Project
University of Japan
Galatea is a project for providing an open-source, license-free software toolkit for building anthropomorphic spoken dialogue agents. In other words, using this toolkit, you can build your own unique life-like visual agent that can communicate with you via spoken language. (note: Anthropomorphic agent is being called in various ways: digital human, virtual human, avatar, animated agent, life-like agent, human-like agent, talking head, animated human image, ... etc. with slightly different emphasis to its various aspects.) - Regulus
NASA
Regulus is a Prolog-based toolkit for building spoken dialogue systems. - The SEMAINE project
The aim of the SEMAINE project is to build a Sensitive Artificial Listener – a multimodal dialogue system with the social interaction skills needed for a sustained conversation with a human user. The system will emphasise “soft” communication skills, i.e. non-verbal, social and emotional perception, interaction and behaviour capabilities. The Sensitive Artificial Listener paradigm involves only very limited verbal capabilities, but has been shown to be suited for prolonged human-machine interaction. In this paradigm, we will build a real-time, robust interactive system perceiving a human user's facial expression, gaze, and voice, and engaging with the user through an Embodied Conversational Agent's body, face and voice. The agent will exhibit audiovisual listener feedback in real time while the user is speaking, and will take the user's feedback into account while the agent is speaking. The agent will pursue different dialogue strategies depending on the user's state; it will learn to interpret the user's non-verbal behaviour and adapt its own behaviour accordingly. - Jaspis
University of Tampere, Finland, Markku Turunen
Jaspis is a framework for adaptive speech applications. Jaspis is designed to support distributed, highly context-sensitive applications that adapt to the user and the environment. It is designed multilingual applications in mind. The architecture is based on the shared information management and on the agent-evaluator-manager -paradigm. - TrindiKit
Göteborg University Dialogue Systems Lab, Sweden, Staffan Larsson
TRINDI (Task Oriented Instructional Dialogue) KIT is a toolkit for building and experimenting with dialogue move engines andinformation states, that has been developed in theTRINDI and SIRIDUS projects, and is currently being developed further in the TALK project. - Midiki - MITRE Dialogue Toolkit,
Midiki is a portable toolkit for building dialogue managers in Java. It implements the information-state model of dialogue as pioneered in Trindikit: a rule-based, theory-neutral, platform agnostic model. - WAMI: Web-Accessible Multimodal Applications
CSAIL, MIT, USA, Alex Gruenstein, Ian McGraw
Wami is an open source Javascript API for speech recognition. This project contains both the server and client side code necessary to host a web-based API for speech recognition. This project does not contain a speech recognizer. This project provides all the plumbing you'll need to give a speech recognizer a web-interface. There's even an iPhone app associated with it (search "wami" in the app store). - CSLU Toolkit
Center for Spoken Language Understanding (CSLU) OGI Campus, Oregon Health & Science University (OGI/OHSU), John-Paul Hosom
A Platform for Research and Development of Spoken-Language Systems
The CSLU Toolkit was created to provide the basic framework and tools for people to build, investigate and use interactive language systems. These systems incorporate leading-edge speech recognition, natural language understanding, speech synthesis and facial animation technologies. The toolkit provides a comprehensive, powerful and flexible environment for building interactive language systems that use these technologies, and for conducting research to improve them.
Because of the Toolkit's easy to use graphical authoring tools, it has spread from the halls of higher education into homes, primary and secondary schools. People who are currently using the Toolkit range from fifth-graders who create school projects to university professors who are experimenting with core language technologies and spoken dialogue systems. - Ariadne
Matthias Denecke
Ariadne is an open source spoken dialogue system developed as a thesis research to prove a generic task and language independent dialogue system that generates interactions with users based on a description of an application.
The system architecture is a layered microkernel architecture which means than that you can replace pretty much every component other than the central dialogue processing component. This includes parsers, natural language generation, the mechanisms of the underlying logics and so on. The price to pay is that if you want to do that you will have to stick to the APIs provided.
It also means that you can develop entire dialogue systems without the need for complicated middleware, socket connections and servers. This should make it easier to port this system to embedded applications. - DIPPER
Dialogue Prototyping Equipment & Resources,
DIPPER is a collaborative initiative of the Language Technology Group at The University of Edinburgh and CSLI Stanford to create a shared software database to comfort in prototyping dialogue systems, using OAAas communication protocol. DIPPER is not a dialogue system itself, but DIPPER supports building (spoken) dialogue systems, by offering interfaces to speech recognisers (Nuance), speech synthesisers (Festival), dialogue managers, natural language understanding, and automated reasoning (SPASS, Bliksem, Mace, PARADOX). - PED: A Planner for Efficient Dialogues
University College Dublin, Ireland, Bryan McEleney
PED is a dialogue management system that uses a probabilistic nested belief model to choose dialogue strategies from a Bayesian game tree. The system designer is only required to supply a context-free dialogue grammar with preconditions. Using this grammar, PED constructs game trees (like the one below) to represent the outcomes of the dialogue. PED uses a belief revision function that makes inferences from the acts observed in the dialogue to update the belief model. PED is more efficient than other systems because it uses a probabilistic belief model. However, the belief revision function described here is approximate and I have not yet seen its solution. - Open Allure Dialogue System
Open Allure is a project aimed at developing new ways to share what we know with one another by permitting the collaborative creation and experience of interactive dialogs. These verbal exchanges give your interaction with the computer a very different quality and permit immediate feedback to help reinforce or reorganize your thinking. Because voice recognition is still imperfect, the interface could also support making choices by gesture: your webcam watches you as you raise your hand.
So with the same basic hardware (headset microphone and webcam) you might use with Skype, Open Allure offers you a conversation with your computer: it talks, it listens, it watches, it responds.
As of 2011, Open Allure is multilingual and mobile. Portuguese (pt) and Italian (it) versions are available for download. Please request other languages if you have interest. An Android version and Java source code is also available. - OWLSpeak, an ontology-based spoken dialogue management system
Uml University, Germany
The system is written in Java and uses OWL Spoken Dialogue Ontologies to generate VoiceXML dialogue documents. OwlSpeak was implemented within the framework of the EU-funded project ATRACO - Adaptive and TRusted Ambient eCOlogies.
References:
- Dynamic use of Ontologies to enhance Dialogue Systems
- Robots@L2F
- www.infj.ulst.ac.uk/sdt/
- www.infj.ulst.ac.uk/~cbdg23/book/projects.htm
- www.infj.ulst.ac.uk/~cbdg23/book/resources.htm
- www.cs.cmu.edu/~dbohus/SDS/index.html
- www.ling.gu.se/~sl/dialogue_links.html
- liceu.uab.es/~joaquim/speech_technology/tecnol_parla/dialogue/refs_dialeg.html
- www.speech.kth.se/multimodal
- mambo.ucsc.edu/psl/fan.html
- www.cs.uta.fi/hci/spi/applications.html
- www.speech.kth.se/higgins
- www.cs.cmu.edu/~usi
- www-csli.stanford.edu/semlab-hold/witas
That's it! Hope this can be useful...
