![]() |
June 5, 2023 | ||
Brainhat is programmed in English. This is the program for the audio clip on the home page:
The interaction of the audio clip is simply a continuation of the
original programming.
The subject areas are segmented into individual
sub-domains or memes.
You can see the separations highlighted in the
comments of the code above.
Taken together, the memes are bundled into
a memeplex or mplex.
Brainhat can shift focus of the conversation to the meme which
seems most appropriate at a given time.
We included some valley girl-isms (e.g. "totally...") so that
you can hear the shifts in progress.
Notice that we can take the initiative in the conversation away from the
program.
As an example, we answered Brainhat's question ("are you a guy?") with
a question of our own.
At any time, we can return to an earlier point in the discussion as was
demonstrated when we asked the program to recall the speaker's name.
Whole conversations can be stored and restored intact via a SQL back-end.
This clip was taken with Dragon NaturallySpeaking Version 6 and
Rhetorical Systems "Valley Girl" (USF012).
Note: speech recognition is a challenge.
It requires a well-trained engine to duplicate the performance of this
clip.
Copyright © 1996-2009, Kevin Dowd
|
|