Well, Ann, that's a mouthful you've asked, and given your deadline, I'm not
sure I can answer all the questions (maybe others will have ideas), but I'll
have a go with a couple of things.
1.) Do you agree with the conclusion that the end user needs to have available
all 3 products within this 30 day loan period?
AT the danger of angering Big Blue, I don't really see any strong need (other
than fair play) to present to present ViaVoice. It's only real advantage over
NaturallySpeaking is its cute little talking head for text-to-speech readback.
And at any rate, I think that a person needs to devote 30 days for training
and giving a reasonable trial to just one of the continuous speech products.
Having 2 to test in 30 days would be excruciating I think. Having to test
DragonDictate in that same time frame could also be difficult, as well, but
less so I think, since it operates differently (and it also can operate in
conjunction with NatSpeak).
On top of the above, having done a fair amount of teacher training and tech
transfer work, I'm also a bit concerned about the likelihood of getting good
results / trials with this technology without substantial support and
training. The materials that come with the products are weak, and resellers
(and clinics like ours at Children's Hospital) devote a fair amount of time to
providing training for speech recognition.
2a.) If no- why? What other method do you see as viable?
Well, our grant is working on the training materials, but we see their
development as a long term project. A (not so) quick and (not) dirty training
demo video, edited in the can, might give people SOMETHING to work from. We
did that once for a piece of word prediction software we created, and it was
not video store material, but it worked pretty well, and was produced quickly
and at little expense than our own time.
2.b) If yes, can they all be loaded on to one (1) PC laptop (233 mhz mmx
Pentium processor with 64 megs of RAM and soundblaster sound card 16 bit
sound)? Would there be conflicts in just loading all three applications on
one machine (drivers, soundcard settings etc.) We have to make this as
problem free as possible- keeping it simple.....
Yes, all 3 seem to work fine residing on the same machine, and I've even had
DDW and VV open at the same time, but I don't think that's smart as a rule. As
I said above, DDW and NatSpeak can now operate together, so that's no problem.
re: Qs 3 & 4, I've talked about them a bit above. Maybe others have ideas.
5.) MAC> Power Secretary: Is this even worth loading onto the MACS if they are
configured for specific LD's relating to written language- if PC's are not
available or other software needed is available in MAC format only? I have
heard no discussion on this product what so ever and was told by DRAGON that
they have NO intention of ever coming out with new or improved PW for
MAC......
I have always been a Mac lover, but (and it pains me to say this), I see no
reason to use Power Secretary - it is old technology and it feels quite clunky
now. There is no other compelling reason regarding Mac-only software to
consider PowerSec. Many schools of course have only Macs, and that can be an
issue, but otherwise, the poor Mac is really dead in the speech recognition
game.
6.) I would like to develop materials to go out with different configurations>
what would YOU put in the VR guide as tips for end users? I have saved some
of your troubleshooting comments from emails> would y'all mind if I use them
in this handbook?
If they are statements from the spk2wrt website, they can be used but I think
they need to be attributed as from the website, owned by Education Development
Center, Newton, MA.
Best of luck and let us know what happens, what works, what doesn't, etc.
Bob Follansbee
This archive was generated by hypermail 2b29 : Tue Jan 04 2000 - 12:43:30 EST