Speech interfaces are commonplace in many types of robots and robotic applications. Despite the progress in speech recognition and many other areas of natural language processing in recent years, failures of speech interfaces in robotic scenarios are numerous and commonplace. In contrast to the common experience of failure of speech interfaces in robotics, the literature is positively skewed towards the success and good performance of these. The documentation of failure speech interfaces on the other hand is exceedingly rare and a systematic study of failure and its causes is non-existent.

The WTF workshop would like to address this gap by bringing together a multidisciplinary group of researchers from the fields such as robotics, human-robot Interaction (HRI), natural language processing, conversation analysis and pragmatics. The central idea of the workshop is to discuss the multitude of failures of speech interfaces openly and, if possible, systematically, in the hope of identifying the most fruitful directions for overcoming these failures in future systems. The envisioned types of failures may range from failures of speech recognition to pragmatic failures and infelicities.

The workshop will focus on human-robot joint action scenarios involving multimodal coordination between humans and robots as these are the norm in scenarios where robotic speech interfaces are deployed.

We also intend to identify a prototypical application scenario embodying most of the identified challenges and that may act as a future benchmark allowing the measurement of progress of speech interfaces in multimodal HRI.

Sponsors

This workshop in sponsored by the UK-RAS network: https://www.ukras.org.uk/