United States | Worldwide
Go
IP TelephonyContact CentersUnified CommunicationsCEBP
Feng LiuFeng Liu
Research Scientist

"The great theme about research is to turn fantasies into reality""
My Research
My research focuses mainly on two areas. One is human-machine interaction, including speech recognition, natural language understanding, multimedia multimodal dialogue systems, and natural language call routing (NLCR). Another one is on web services and service-oriented architecture (SOA) in converged communications systems, including voice over IP (VoIP), web services, and its applications in converged communication systems.
Projects
Web Services and Communication Service Enablement
This project is on web service enablement framework/middleware, e.g. two-way web service application proxy (2SAP) that supports sync/async request and response, event notification, and session-based web services. Our session based web service solution is contributed to the standard body ECMA International, and it is adopted as an ECMA International standard ECMA-366 (WS-Session).
 
LiveWire with IBM’s Extreme Blue program
We engaged and provided IBM team with 2SAP based web service enabled Avaya Meeting Exchange conference services. A prototype system was built and applications developed that integrates Avaya’s conferencing services, IBM LDAP blue page directory web services, presence web services, etc. into Lotus Notes. It makes Avaya Meeting Exchange conference services generally available for IBM WAS 6.0 applications through WSBPEL. It was demonstrated at IBM Expo with rave comments.
 
Web Service SIP (WSIP)
WSIP is a new paradigm of communication over IP. Each WSIP node is both a SIP endpoint that communicates through SIP signaling, and a web service (SOAP) node that provides a native and generic service integration environment for binding SIP based communication in web service. We developed a prototype WSIP, and the overlay of Web Services and SIP on top of existing IP substrate provides a new converged communication paradigm over IP.
 
Web Service SIP (WSIP)
WSIP is a new paradigm of communication over IP. Each WSIP node is both a SIP endpoint that communicates through SIP signaling, and a web service (SOAP) node that provides a native and generic service integration environment for binding SIP based communication in web service. We developed a prototype WSIP, and the overlay of Web Services and SIP on top of existing IP substrate provides a new converged communication paradigm over IP.
 
Performance of voice portal services in VoIP environments
The research focus of this project is on the adverse effect of delay, jitter and packet loss for voice portal services over IP, e.g. the performance degradation issues of the DTMF detection in voice portal under adverse channel conditions over IP.
 
Natural Language Call Routing (NLCR)
NLCR is related to natural language understanding, and the goal is to determine the caller’s intension and transfer the caller to the desired destination through natural language based dialogue interaction. We developed an approach that uses word-cluster terms in NLCR, which is more robust than word terms. To combine both word cluster and word terms information to enhance the robustness and performance in NLCR, we proposed an algorithm to select the suitable terms using the information theoretical approach based on information gain (IG).
 
MTIP
The research of this project is to research on multimodal multimedia human-computer interaction system. This is an ongoing research project, and involves all aspects of the dialogue systems. We developed an extensible three-tier XML dialogue system architecture. It provides a generic and layered XML abstraction to support different dialogue XML based applications. We developed the Extensible Micro-Flow (XM-Flow) framework based on a flow extension of SMIL and the extensible semantic model of EMMA for simultaneous multimodal interaction.
 
WMTIP
Wireless multimodal multimedia dialogue systems based on wireless web and dialogue system convergence over converged networks. In WMTIP, mobile users can interact with the dialogue system MTIP through voice and web to request services, and it delivers the advanced multimodal and multimedia content () to mobile users, including voice, text, picture, table, form, and even active links. The integration of WAP push/pull method in multimodal interaction provides a new paradigm of wireless access of enterprise communication services.
 
Hybrid-VoiceXML for multimodal dialogue systems
We developed a distributed multimodal dialogue system architecture based on the concept of hybrid-VoiceXML. It utilizes a special hybrid-construct to integrate multiple multimedia, multimodal processes into one dialogue that includes VoiceXML as its voice modality. The hybrid-construct in our approach has several important functions. It can be used to form platform independent and distributed extensions for multimodal dialogue interaction beyond voice.
 
Low cost speech recognition engine
We developed an ASR engine that provides low cost and high-performance ASR system for CRM applications and Avaya’s natural language call steering systems.
 
 
 
Publications
Selected Publications
W. Chou, L. Li, and F. Liu, “Web Service Enablement of Communication Services”,  Proceedings of IEEE International Conference on Web Services (ICWS’05), pp.393 – 400, July 2005.
 
L. Li, F. Liu, and W. Chou, “An Extensible Three-Tier XML Dialogue System Architecture for Multimodal Interaction and Automated Agent Services”, International Conference on Internet and Multimedia Systems, and Applications (IMSA 2005).
 
W. Chou, L. Li, F. Liu, “Web Service for Communication Service Management”, The 17th Int. Conf. on Software Engineering and Knowledge Engineering (SEKE), pp. 584-589, July 2005.
 
F. Liu, W. Guo, W. Chou and E. Wong, An Approach of Integrating SIP in Converged Multimodal/Multimedia Communication Services, the Journal of Telecommunication Systems. 28:3, 4, pp387-405, 2005
 
M. Afify, F. Liu, H.Jiang, and O. Siohan, “A new verification-based fast match for large vocabulary continuous speech”, IEEE Trans. On Speech And Audio Processing. Vol. 13, no. 4, July 2005, pp.546-553
 
W. Chou and F. Liu, Voice Portal Services in Packet Network and VoIP Environment , ICSLP 2004
 
F. Liu, J. Li, W. Chou, and X. Shan, “A n architecture of wireless web and dialogue system convergence for multimodal service interaction over converged networks”, Proc. COMSAC2003, Dallas, TX, November, 2003
 
F. Liu, W. Chou, L. Li and J. Li, WSIP – Web Service SIP Endpoint for Converged Multimedia/Multimodal Communication over IP, IEEE International Conference on Web Services (ICWS’2004)
 
L. Li, F. Liu, and W. Chou, “An Information Theoretic Approach for Using Word Cluster Information in Natural Language Call Routing”. EuroSpeech 2003, Sept. 1-4, Geneva, Switzerland
 
F. Liu, A. Saad, L.Li, W.Chou, “A Distributed Multimodal Dialogue System Based On Dialogue System And Web Convergence”, ICSLP 2002.
 
F. Liu, M. Afify, H. Jiang, and O. Siohan, A new verification-based fast match approach to large vocabulary continuous speech recognition, In Proceedings of European Conference on Speech Communication and Technology, (Eurospeech2001), Aalborg, Denmark, September 2001
 
 
 
 
CONTACT INFORMATION
Phone908-695-5156
letterfliu1_AT_avaya_DOT_com
Connect
with Avaya