Data-Driven Methods for Adaptive Spoken Dialogue Systems

1. Front Matter

   Pages i-ix

   PDF

2. Conversational Interfaces

   • Oliver Lemon

   Pages 1-4

3. Developing Dialogue Managers from Limited Amounts of Data

   • Verena Rieser, Oliver Lemon

   Pages 5-17

4. Data-Driven Methods for Spoken Language Understanding

   • James Henderson, Filip Jurčíček

   Pages 19-38

5. User Simulation in the Development of Statistical Spoken Dialogue Systems

   • Simon Keizer, Stéphane Rossignol, Senthilkumar Chandramohan, Olivier Pietquin

   Pages 39-73

6. Optimisation for POMDP-Based Spoken Dialogue Systems

   • Milica Gašić, Filip Jurčíček, Blaise Thomson, Steve Young

   Pages 75-101

7. Statistical Approaches to Adaptive Natural Language Generation

   • Oliver Lemon, Srini Janarthanam, Verena Rieser

   Pages 103-130

8. Metrics and Evaluation of Spoken Dialogue Systems

   • Helen Hastie

   Pages 131-150

9. Data-Driven Methods in Industrial Spoken Dialog Systems

   • Roberto Pieraccini, David Suendermann

   Pages 151-171

10. Conclusion and Future Research Directions

    • Olivier Pietquin

    Pages 173-177

Data driven methods have long been used in Automatic Speech Recognition (ASR) and Text-To-Speech (TTS) synthesis and have more recently been introduced for dialogue management, spoken language understanding, and Natural Language Generation. Machine learning is now present “end-to-end” in Spoken Dialogue Systems (SDS). However, these techniques require data collection and annotation campaigns, which can be time-consuming and expensive, as well as dataset expansion by simulation. In this book, we provide an overview of the current state of the field and of recent advances, with a specific focus on adaptivity.

• Adaptive user interfaces
• Machine Learning
• Spoken Dialogue Systems
• User Simulation

• , Mathematics and Computer Science, Heriot Watt University, Edinburgh, United Kingdom

  Oliver Lemon

• , Metz Campus - IMS Research Group, SUPELEC, Metz, France

  Olivier Pietquin

Oliver Lemon is a Reader and head of the Interaction Lab in the school of Mathematical and Computer Sciences at Heriot Watt University, Edinburgh. Dr. Lemon is currently serving as the Program Chair for SIGDial 2010 and as a member of the Program Committee of INLG 2010. He is also on the Editorial Board of the new journal "Dialogue & Discourse". Prof. Pietquin and Dr. Lemon were co-chairs of the special session "Machine learning for adaptivity in spoken dialogue systems" at the InterSpeech 2009 conference, which inspired the development of this book.

Olivier Pietquin is an Associate Professor at the Ecole Superieure d'Electricite (Supelec, France), where he founded and currently heads the "Information, Multimodality & Signal" (IMS) research group. He is an elected member of the IEEE Speech and Language Technical Committee. Prof. Pietquin has four patents and has been published in over 45 journal articles, edited books, and conference proceedings.

Policies and ethics