Tanvi Dinkar

Research Associate in the Interaction Lab at Heriot Watt University

Session: The human side of conversational AI

Tanvi Dinkar is a Research Associate in the Interaction Lab at Heriot Watt University.  She works on safety and robustness for Conversational AI with Prof. Verena Rieser and Prof. Ekaterina Komendantskaya. She recently completed her PhD in Computer Science at Télécom Paris. In her PhD, she studied the representations of disfluencies for Spoken Language Understanding (SLU), and how they can be modelled as informative signals of communication, rather than simply being removed as noise. During her PhD, she was a Marie Curie Early Stage Researcher at ANIMATAS. Her research interests include safety and robustness in Conversational AI, SLU, Psycholinguistics, and the discrepancies between the way that people speak versus the way that people write. Prior to this, she was a dialogue engineer at Nuance (now Microsoft), coding dialogue systems for the automotive industry. She decided to pursue research when she saw from customer tickets that the task-oriented dialogue systems were not robust to people speaking naturally. She has two masters from the University of Edinburgh, one in Linguistics and one in Speech and Language Processing. Once upon a time, she completed an undergraduate degree in Journalism and Literature.