Preface
Acknowledgments
Glossary
Introducing Dialogue Systems
What is a Dialogue System?
Why Develop a Dialogue System?
A Brief History of Dialogue Systems
Text-Based and Spoken Dialogue Systems
Voice User Interfaces
Chatbots
Embodied Conversational Agents
Robots and Situated Agents
Limitations of Early Dialogue Systems
Present-Day Dialogue Systems
Dialogue Systems on Messaging Platforms
Dialogue Systems on Smartphones
Dialogue Systems on Smart Speakers and Other Devices
Dialogue Systems in Cars
How Current Dialogue Systems Are Different
Modeling Conversation in Dialogue Systems
User-Initiated Dialogues
System-Directed Dialogue
Multi-Turn Open-Domain Dialogue
Designing and Developing Dialogue Systems
Rule-Based Dialogue Systems: Architecture, Methods, and Tools
A Typical Dialogue Systems Architecture
Automatic Speech Recognition (ASR)
Natural Language Understanding (NLU)
Dialogue Management
Natural Language Generation (NLG)
Text-to-Speech Synthesis (TTS)
Designing a Dialogue System
Tools for Developing Dialogue Systems
Visual Design Tools
Scripting Tools for Handcrafting Dialogue Systems
Advanced Toolkits and Frameworks
Research-Based Toolkits
Which is the Best Toolkit?
Rule-Based Techniques in Dialogue Systems Participating in the Alexa Prize
Statistical Data-Driven Dialogue Systems
Motivating the Statistical Data-Driven Approach
Dialogue Components in the Statistical Data-Driven Approach
Natural Language Understanding
Dialogue Management
Natural Language Generation
Reinforcement Learning (RL)
Representing Dialogue as a Markov Decision Process
From MDPs to POMDPs
Dialogue State Tracking
Dialogue Policy
Problems and Issues with Reinforcement Learning and POMDPs
Evaluating Dialogue Systems
How to Conduct the Evaluation
Laboratory Studies vs. Evaluations in the Wild
Evaluating Task-Oriented Dialogue Systems
Quantitative Metrics for Overall Dialogue System Evaluation
Quantitative Metrics for the Evaluation of the Sub-Components of Dialogue Systems
Qualitative/Subjective Evaluation
Evaluating Open-Domain Dialogue Systems
Evaluation at the Level of the Exchange
Evaluation at the Level of the Dialogue
ChatEval: A Toolkit for Chatbot Evaluation
Evaluations in Challenges and Competitions
Evaluation Frameworks
PARADISE
Quality of Experience (QoE)
Interaction Quality
What is the Best Way to Evaluate Dialogue Systems?
End-to-End Neural Dialogue Systems
Neural Network Approaches to Dialogue Modeling
A Neural Conversational Model
Introduction to the Technology of Neural Dialogue
Word Embeddings
Recurrent Neural Networks (RNNs)
Long Short-Term Memory Units
The Encoder-Decoder Network
Retrieval-Based Response Generation
Task-Oriented Neural Dialogue Systems
Open-Domain Neural Dialogue Systems
Alexa Prize 2020
Google's Meena
Facebook's BlenderBot
OpenAI's GPT-3
Some Issues and Current Solutions
Semantic Inconsistency
Affect
Dialogue Systems: Datasets, Competitions, Tasks, and Challenges
Datasets and Corpora
Competitions, Tasks, and Challenges
Additional Readings
Challenges and Future Directions
Multimodality in Dialogue
Multimodal Fusion
Multimodal Fission
Multimodality in Smartphones and Smart Speakers with Displays
Visual Dialogue and Visually Grounded Language
Data Efficiency: Training Dialogue Systems with Sparse Data
Knowledge Graphs for Dialogue Systems
Reasoning and Collaborative Problem Solving in Dialogue Systems
Discourse and Dialogue Phenomena
Making Reference
Detecting, Maintaining, and Changing Topic
Multi-Party Dialogue
Incremental Processing in Dialogue
Turn-Taking in Dialogue
Hybrid Dialogue Systems
Dialogue with Social Robots
Dialogue and the Internet of Things
Social and Ethical Issues
The Way Ahead
Toolkits for Developing Dialogue Systems
Bibliography
Author's Biography