Ben, Dr. Eliza with the Gracie interface to Dragon NaturallySpeaking makes a really spectacular speech I/O demo - when it works, which is ~50% of the time. The other 50% of the time, it fails to recognize enough to run with, misses something critical, etc., and just sounds stupid, kinda like most doctors I know. Even when it fails, it still babbles on with domain-specific comments.
Results are MUCH better when a person with speech I/O and chronic illness experience operates it. Note that Gracie handles interruptions and other violations of conversational structure. Further, it speaks in 3 voices, one for the expert, one for the assistant, and one for the environment and OS. Note that the Microsoft standard speech I/O has a mouth control that moves simultaneously with the sound, that is pasted on an egghead face, so you can watch it speak. Note that the speech recognition works AMAZINGLY well, because the ONLY thing it is interested in are long technical words and relevant phrases, and NOT in the short connecting words that are what usually gets messed up. When you watch what was recognized during casual conversation, what you typically see is gobbledygook between the important stuff, which comes shining through. There are plans to greatly enhance all this, but like everything else on this forum, it suffers from inadequate resources. If someone is looking for something that is demonstrable right now to throw even modest resources into... That program was then adapted to a web server by adding logic to sense when it was on a server, whereupon some additional buttons appear to operate and debug it in a server environment. That adapted program is now up and running, without any of the speech I/O stuff, on http://www.DrEliza.com. I know, it isn't AGI, but neither is anything else these days. Any interest? Steve ============ On Sat, Aug 7, 2010 at 6:10 PM, Ben Goertzel <b...@goertzel.org> wrote: > Hi, > > A fellow AGI researcher sent me this request, so I figured I'd throw it > out to you guys.... > > **** > I'm putting together an AGI pitch for investors and thinking of low > hanging fruit applications to argue for. I'm intentionally not > involving any mechanics (robots, moving parts, etc.). I'm focusing on > voice (i.e. conversational agents) and perhaps vision-based systems. > Hellen Keller AGI, if you will :) > > Along those lines, I'd like any ideas you may have that would fall > under this description. I need to substantiate the case for such AGI > technology by making an argument for high-value apps. All ideas are > welcome. > **** > > All serious responses will be appreciated!! > > Also, I would be grateful if we > could keep this thread closely focused on direct answers to this > question, rather than > digressive discussions on Helen Keller, the nature of AGI, the definition > of AGI > versus narrow AI, the achievability or unachievability of AGI, etc. > etc. If you think > the question is bad or meaningless or unclear or whatever, that's > fine, but please > start a new thread with a different subject line to make your point. > > If the discussion is useful, my intention is to mine the answers into a > compact > list to convey to him > > Thanks! > Ben G > > > ------------------------------------------- > agi > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/ > Modify Your Subscription: > https://www.listbox.com/member/?&; > Powered by Listbox: http://www.listbox.com > ------------------------------------------- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244&id_secret=8660244-6e7fb59c Powered by Listbox: http://www.listbox.com
