![[Univ of Cambridge]](http://www.eng.cam.ac.uk/images/house_style/uniban-s.gif){kind=small}
![[Dept of Engineering]](http://www.eng.cam.ac.uk/images/house_style/engban-s.gif){kind=small}
|
|
|
Jason D. WilliamsI've moved: I'm now with AT&T Labs - Research.
I'm a Ph D student in the Engineering Department, University of Cambridge. I'm in the Speech Research Group of the Machine Intelligence Laboratory in the Information Engineering Division. I joined the group in October, 2002, and I'm also a member of Churchill College. My Ph D advisor is Prof Steve Young.
I received an undergraduate (B.S.E.) degree in Electrical Engineering from Princeton University, and an M Phil (masters) degree in Computer Speech and Language Processing from University of Cambridge. I subsequently worked for McKinsey (Associate) in Palo Alto, California. I then worked for Tellme in Mountain View, California and also in Brussels, Belgium. Prior to starting my Ph D, I worked full-time for Edify in London.
A (*) indicates the work was performed other than at CUED.
Dialogue Management using Partially Observable MDPs. Talk for
"Dialogs on Dialogs" discussion
group (Carnegie Mellon University). 3 June 2005. [pdf]
Probabilistic models of human/computer dialog. Informal presentation given to "Dialogs on Dialogs" discussion group (Carnegie Mellon University). 23 January 2004. [pdf]
Andrew Shaw, Jason D. Williams.
Automating Call Routing in the
Contact Center. SpeechTEK 2002, 31
October 2002, New York, NY, USA. (*)
Andrew Shaw, Jason D. Williams.
Open Usability: a Case Study. SpeechTEK usability workshop,
"Meeting Business Objectives through Speech Usability," 1 November 2002, New
York, NY, USA. (*)
Andrew Shaw, Jason D. Williams. Usability in the Discovery and Early Development Phase. SpeechTEK 2002, 1 November 2002, New York, NY, USA. (*)
Jason D. Williams, Pascal Poupart, and Steve Young.
Using Factored Partially Observable Markov Decision Processes with Continuous Observations for Dialog Management. Cambridge University Engineering Department Technical Report: CUED/F-INFENG/TR.520. 24 March 2005. [pdf]The SACTI-1 Corpus is a corpus of 144 human-human dialogues in the tourist information domain using the Simulated ASR Channel.
The paper "Characterizing Task-Oriented Dialog using a Simulated ASR Channel" above gives a brief description and an analysis of the corpus.
The paper "A Framework for Wizard-of-Oz Experiments with a Simulated ASR-Channel" above describes the Simulated ASR Channel used to collect the corpus.
The tech report "The SACTI-1 Corpus: Guide for Research Users" below gives complete detail of the content of the corpus.
For questions regarding access
to the corpus, please email me:
jdw30@cam.ac.uk.
Jason D. Williams and Steve
Young. The SACTI-1 Corpus: Guide for Research Users.
Cambridge University Department of Engineering Technical Report
CUED/F-INFENG/TR.482.
February 2005. [pdf]
Karl Weilhammer, Jason D.
Williams, and Steve Young. The SACTI-2 Corpus: Guide for Research Users.
Cambridge University Department of Engineering Technical Report
CUED/F-INFENG/TR.505.
February 2005 [pdf]. (For
information about this corpus, please contact Karl Weilhammer).
Jason D. Williams. Toy spoken dialogue
system POMDP: Problem definition. Informal discussion document,
15 December 2003. [pdf]
Jason D. Williams. A Probabilistic Model of Human/Computer Dialogue with Application to a Partially Observable Markov Decision Process. Ph D first year report. Department of Engineering, University of Cambridge. 28 August 2003. [pdf]
| Jason D. Williams Fallside Laboratory Department of Engineering University of Cambridge Trumpington Street, Cambridge CB2 1PZ, United Kingdom |
Email: jdw30@cam.ac.uk |