
Multi-threaded dialogues
We envision that next-generation speech user interfaces (SUI) will enable users to engage
in multiple tasks at the same time in a hands-busy and eyes-busy environment, resulting
in conversations in which threads of dialogues are interleaved. Such a capability would
be useful in a number of ubiquitous computing domains where a subject needs to interact
with multiple devices in real-time. For example, it would allow a police officer to receive
computer-aided dispatches while querying remote license plate databases. It would also
allow a driver in a car to catch up on email and occasionally check upcoming traffic
conditions or receive navigation instructions. In our definition, each dialogue thread is
about a different topic.
The major hypothesis of our research is that we can model the conventions humans use in
multi-threaded dialogues with other humans and develop next-generation SUIs based on
this model.
Return to CATLab R&D Areas |