[NTLK] Speech recognition.

[Forrest Buffenmyer]

Joel:

I used dictation software in the late 90s...and it was okay. While admittedly it should be much better now (I'm sure it's advanced quite a bit), the problem wasn't the accuracy as much as it was what LG said: there are some places where it's just not appropriate.

In 1986, just before I moved here to Arizona, I brought my former fiance on a vacation here to see what it was all about. Among the places we visited was the South Rim of the Grand Canyon...at dawn. We arrived around 6am...it was twilight, and a very different kind of dawn--just the chirps and songs of a few birds, some of whose cries came drifting up out of the canyon in a very interestingly, ethereal, echo-ey kind of way. There was also some rustling sounds just below us...it was very calming and serene.

That is, until the guy with the video camera showed up.

I have no problem with anyone who wants to capture nature on film...as long as they do it QUIETLY. Well, because of the way video cameras were made in those days, with the sort of mini shotgun microphones mounted on the top, to capture any speech on tape you had to yell. And that's what he was doing...yelling, as if trying to talk to someone 50 yards or so away.

"HERE WE ARE AT THE GRAND CANYON...IT'S JUST ABOUT DAWN, AND THERE'S AN EERIE CALM PRESENT."

My thoughts were, "OF COURSE THERE IS! YOUR SHOUTING HAS ALARMED ALL THE WILDLIFE AND THEY HAVE ALL SHUT UP AND RAN AWAY!"

This would clearly NOT be a place where I would want to be dictating my thoughts into some sort of device...as I said before, if it works for you, great! But speaking just for me--I would rather not have to evaluate the situation to determine if I could comfortably use dictation or not.

AND...of course, to go along with LG's comments, I must (again) post the link to this YouTube video: http://www.youtube.com/watch?v=v9kTVZiJ3Uc.

Thanks,
--Forrest

Sent from my First Generation iPad

On Mar 27, 2012, at 2:26 AM, "Joel M. Sciamma" <joel at inventors-emporium.co.uk> wrote:

> It's not about replacing HWR with speech or trying to shoehorn any of these technologies into being a single solution but instead it's about giving you options depending on the task and the context.

And, Lord Groundhog wrote:
<snip>but at
least there's some hope to achieve this one of my childish scifi dreams and
control my electronic devices by speaking to them.


"Computer, send e-mail!"

Hmmm.  Needs work.<snip>