>> Voice recognition is getting more and more accurate by the day, but it is
>> important to realize that (1) it's still a very tough problem to solve and
>> (2) there are parallel efforts attacking the problem. The parallel tracks
>> are handling (1) an unlimited domain for a single speaker and (2)
>> a limited domain for any speaker.
Dave's nailed it on the head.
To add onto Dave's synopsis (I did a bit of looking into this about 25
years, along with vision systems for robots), what Gil is looking for is
actually quite reasonably these days. The tales of woe are generally
from folks who have expectations beyond the current realm of technology.
For a single person to train a system with a distinct vocab ("SQL" is a
language, not the followup to a hit movie), particularly with someone
who knows what they're doing and has the discipline to adapt a bit
(which fits Gil to a 't'), voice recog is definitely a reasonable
vehicle right now.
One gets into trouble if one won't exercise a little discipline
(speaking too fast, background noise, slurring words, using slang, etc.)
And even then, with just a single speaker, systems will learn to adapt,
and translate "y'wl" to the more proper "you-all".
Whil
_______________________________________________
Post Messages to: [email protected]
Subscription Maintenance: http://leafe.com/mailman/listinfo/profox
OT-free version of this list: http://leafe.com/mailman/listinfo/profoxtech
** All postings, unless explicitly stated otherwise, are the opinions of the
author, and do not constitute legal or medical advice. This statement is added
to the messages for those lawyers who are too stupid to see the obvious.
