On 10/07/07, Ovidiu Sabou <ovidiu.sabou(a)gmail.com> wrote:
*Scenario*: one asks a question in natural language
(currently only English)
and the virtual character gives a short answer and provides more information
(the entire Wikipedia article from which the answer was extracted) on a
virtual screen. We use voice recognition and speech synthesis in order to
offer a natural way of obtaining information.
Voice recognition, eh?
Damned if I wasn't thinking about this sort of thing a couple hours
ago while washing up...the vision I had was a device in the home which
would respond to queries, so one could ask, "Wikipedia, how long is
the Great Wall of China", and get an audible response within a
reasonable response time.
That is more or less the extent of the thought, and while there are a
lot of questions and issues to be resolved around it, I see no reason
it couldn't become a reality within a few years.
Rob Church