My research is in statistical methods for Spoken Dialogue Systems, i.e. computer systems with which you can have a conversation. After completing my BA in Mathematics at Cambridge, I obtained an MSc in Speech and Language Processing at the University of Edinburgh, with Colin Matheson supervising my dissertation. Now I am working for a PhD in Cambridge under the supervision of Steve Young in the CUED Dialogue Systems Group.
I have helped develop a couple of online demos for our Restaurant Information dialogue systems. You can call the system using a flash plugin at the group's demo page. There is also a general overview of the BUDS POMDP Dialogue System as it was presented at the NIPS 2012 demo session.
You may read an overview of my research interests as was presented at the Young Researchers Roundtable on Spoken Dialogue Systems in July 2012. [I am on the organising committee for YRRSDS 2013 in Metz.]
Discriminative Spoken Language Understanding Using Word Confusion Networks
Presents a new form of semantic decoding which uses the entire distribution of speech recognition hypotheses to infer the meaning of an utterance. This was presented as a poster at SLT. The dataset used for training the decoder and for the offline evaluation is freely available: In Car SLU Corpus. IEEE Workshop on Spoken Language Technology (SLT), 2012, Miami View PDF Bibtex citation
Recovering from Non-Understanding Errors in a Conversational Dialogue System
A paper stemming from work done in my MSc evaluating a set of strategies a conversational (rather than task-oriented) dialogue system can use to recover from situations where it is unable to understand what the user has asked. The dialogue system concerned is for a robot tourguide which shows visitors around the Edinburgh Informatics Forum. This was an oral presentation at Semdial, and you may view the slides. Workshop on the Semantics and Pragmatics of Dialogue (SemDial), 2012, Paris. View PDF Bibtex citation
N-Best Error Simulation for Training Spoken Dialogue Systems
Policy Optimisation of POMDP-based Dialogue Systems Without State Space Compression
The Effect of Cognitive Load on a Statistical Dialogue System
Statistical Methods for Building Robust Spoken Dialogue Systems in an Automobile