Need help in finding some means of being able to program with speech recognition

  • Thread starter Thread starter Rod
  • Start date Start date
R

Rod

About two weeks ago I had an accident and have broken my left elbow and left
wrist. For doing things like Word or e-mail (I use Outlook for) I have been
using Microsoft's speech recognition and that has been working fine.
However as a professional programmer I need to be able to type. Right now I
can't do that (until my arm heals). I've heard of a program by ScanSoft
called Dragon NaturallySpeaking Professional. I called ScanSoft and talk to
them about Dragon NaturallySpeaking Professional and found that it does not
work with Visual Basic (neither VB6 nor VB.NET) nor C# .NET. So I'm
wondering what other methods of software or whatever there might be able to
use to do my job. Or what agencies that may be available that could help
people in my situation. I would appreciate any help you can give me.



Thank you.



Rod
 
Microsoft Speech Server is the answer, and it's one of the coolest things
I've used. Casey also has some really killer stuff on
www.brains-n-brawn.com that might be of interest. He recently wrote his own
speech rec on the Compact Framework, but his freespeech article does it with
Web Services, Dime and SALT.

This might be of help too http://www.microsoft.com/speech/ and Speech Server
has full Visual Studio .NET integration. I've been blogging about it for a
while http://msmvps.com/williamryan/posts/4250.aspx and I think if you give
it a try, you'll immediately fall in love with it.

HTH,

Bill

--

W.G. Ryan, eMVP

http://forums.devbuzz.com/
http://www.knowdotnet.com/williamryan.html
http://www.msmvps.com/WilliamRyan/
http://www.devbuzz.com/content/zinc_personal_media_center_pg1.asp
 
Rod,

The integration the other poster refers to I believe, is for development of
Speech apps. I do not believe there's anything that will let you use speech
to write C# apps with VS.NET. I may be wrong.

I actually began a project to do precisely this about a year ago and what
there is of it (quite a bit, actually), is sitting on a laptop hard drive
that failed. The data is recoverable but I won't be able to recover it for a
few weeks (need to busy some parts).

So, I guess the answer is that there's not really anything for you that I'm
aware of.

Unfortunately, doing something like voice dictation in notepad won't work
either. There are a great number of hurdles to using voice recognition to
develop apps in VS.NET and I haven't yet figured out the best way to handle
all of them. For example, how do you handle variable names that aren't
pronounceable? Spell them out? How do you handle capitalization in the
following property example

public Font Font
{
get
{
return font;
}
}

The capitalization stuff is a real pain.

I was still largely testing different methods for dealing with these kinds
of issues.

The speech recognition grammar is also quite complex because it has to, to
some degree, reflect the language grammar, which isn't very simple.

Anyway, I guess you probably didn't need to know that much about the issues,
but hey, when you get your arm working again, if you feel like helping me
finish the program, send me an e-mail. Hopefully I'll have the hard drive
recovered by then.

Pete
 
Hello Bill,

I've been looking it's the server after reading your post. I am confused as
to how it is to be used. It looks like it would be something used by call
centers but I am not sure it would be used to help someone program. Can you
please give me a website detailing someone actually having used the Speech
Server to write programs such as VB.NET programs or C# programs.

Thank you .

Rod
 
Rod, the breakup of the Speech SDK is a bit confusing at the moment. MS has
a couple of versions going right now (and I haven't looked in about 6
months, so I'm a bit confused as well).

You have the Speech Application SDK (SASDK) which is an ASP.NET based speech
SDK. Since VS.NET is not web based, that won't do you any good.

Then they have the Speech SDK or Speech API (SAPI), version 5.1 is current
and that's what you would use.

SAPI 5.1 is not a .NET implementation, but is usable from .NET via
automation. I believe the next version of SAPI (6.0??) will be .NET based.
One big difference between SAPI and SASDK is that the grammar format has
changed and I believe SAPI 6.0 will use a grammar format that's compatible
with SASDK. The newer grammar format is nicer, I think and more powerful.

Pete
 
About two weeks ago I had an accident and have broken my left elbow and left
wrist. For doing things like Word or e-mail (I use Outlook for) I have been
using Microsoft's speech recognition and that has been working fine.
However as a professional programmer I need to be able to type.
Click to expand...

I doubt you'll find any speech system for programming
(assuming one exists) any easier to use than typing with your nose.
Since you've got one good hand, you might try one of the one-handed
keyboards. A google for "one handed keyboard" turned up a google of
results. They're not cheap, but cheaper than Dragon Naturally
Speaking. And once you've learned it, you can type while driving when
your arm heals. I've heard some people achieve near the same typing
speed with one hand that they do with both hands, though I can't say I
know anyone who's ever used one.
Also you might try looking for a program which redefines keys
on your keyboard so that you can easily type with one hand. There's
also the "On-Screen Keyboard" that comes with all versions of Windows
(under accessibility). You might also just try a smaller keyboard and
do hunt and peck typing.
 
Back
Top