Natural spoken conversational interfaces have been the long term goal for AI researchers for many decades. Recently, deep learning has facilitated huge improvements in the quality of supervised learning for speech recognition and synthesis, combined with recent advances in reinforcement learning. These trends taken together have finally brought the conversational interface from the lab and into the office, home and even car. These advances make use of many different machine learning algorithm types, and typically rely on multiple components being brought together smoothly. As a result, one of the biggest challenges for the professional is understanding the range of techniques and platforms that are available and how they relate to specific business goals.

Through code examples, discussion and interaction with working systems, this masterclass will guide the professional through the range of models, techniques and frameworks that come together in building conversational agents of different intelligence levels. The course will allow the professional to understand the role of key components in the conversational interface, and to develop the techniques to apply these components to develop concrete working applications.

Dr. Robert Ross is a Senior Lecturer in the School of Computer Science DIT. Robert has a BSc and MSc by Research from University College Dublin, and a PhD from the University of Bremen in Germany. Robert’s research interests are primarily in the areas of Dialogue Systems and more recently in the application of machine learning and deep learning techniques to unstructured data including text, sensor data streams and video. Robert is a Funded Investigator in the SFI funded ADAPT Research Centre and a Principle Investigator in the EI Funded CeADAR Centre.