I think we might have had this conversation before. Damn that flaky grey
matter.
A human to do this job would work well, I guess, but why make them repeat
what you want to say? They'd just carry some kind of speaker, from which
your voice could eminate. Also, I guess it would be a little less
dehumanising if you just asked them to move about in certain ways, rather
than driving them around with a joystick!
Robotics doesn't seem to be up to the job quite yet. However, given a little
time, it could start to come together. Eventually, hiring a robot would have
to be cheaper and offer better control, than hiring a person.
Why aren't there telepresence operations already? The teleoperative would
most likely be a good actor or other skilled communicator, perhaps skilled
enough to do continuous conference-style language interpreting (that'd cost
extra!), with a wearable computing rig. I would think there could be a
market for this, and it might be quite a weird & cool gig for otherwise
unemployed acting types. Better than waiting on tables.
Anyone want to set it up?
Emlyn
----- Original Message -----
From: "phil osborn" <philosborn@hotmail.com>
To: <extropians@extropy.org>
Sent: Saturday, September 02, 2000 4:20 PM
Subject: Re: Teleoperation [was Re: TV: Documentary Science of Beauty]
>From: Ken Clements <Ken@Innovation-On-Demand.com>
>Subject: Teleoperation [was Re: TV: Documentary Science of Beauty]
>Date: Fri, 25 Aug 2000 14:04:17 -0400
>
>Emlyn O'Regan wrote:
>
> >
> > For instance, I'd love to be able to rent a robot body for the day from
>a US
> > robot-rental dealer (robo-hurtz), and be able to turn up to my US
>clients
> > and actually interact with them, whilst tucked in my little hidey-hole
>in
> > Melbourne. I'd pay money for that.
> >
Why waste all that money? You could hire a real, live human being to carry
around an RF linked webcam - or two, for binocular vision - and microphones.
With a fairly simple switching system, you could give them instructions,
or even incorporate a joystick, mouse, or touchpad controller that would use
various input modalities - buzzers, tones, or actual vocal interpretations -
to instruct them how to move, what to look towards, etc.
Your voice would either be a speaker mounted as close to the person's mouth
as possible - probably low neck area, or their voice as an interpreter, with
you hearing and/or reading auto-translation via voice recognition and
translation software. A further check on accuracy could be provided via a
heads-up display for your telepersona operator, who would see the original
voice signal looping back as a retranslation into their own language.
At shows like CES or COMDEX, there are companies that provide professional
generic salespersons to man your booth, which is especially handy for
foreign companies that may not have very many good English speakers
available. This would be a natural extension of their services.
(I first suggested this I think when the Indy came out with decent built-in
voice recognition. My idea then was to take someone who didn't speak the
language being used, take the voice input from other speakers in a
conversation, run it through one of the smarter versions of Ractor, and then
send the Ractor output back into an earphone of the humbot, who would be
repeating it verbatim, not speaking the language. I thought it would be
interesting to see how long it took the other members of the conversation to
notice something odd in this variation on the Turing test.)
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
Share information about yourself, create your own public profile at
http://profiles.msn.com.
This archive was generated by hypermail 2b29 : Mon Oct 02 2000 - 17:36:56 MDT