John Doe writes:

I have a very great demand for that.

As you pointed out to me, you may not be representative.

My system, probably no better than a current store-bought computer,
is running it just fine, input and output.

If it's fast enough, it should.

That depends on what you're talking about. If you're talking about
text to speech, it's very understandable even though probably not
very good sounding to most people. If you're talking about speech
recognition, yes it can be very frustrating.

I'm talking about both. The slowness and lack of accuracy of speech
systems holds them back. That's why people tend not to use them
unless they have to. I'd much rather type than have to speak to my
computer to write things. It would take forever with speech.

Windows supports keyboard input all the way to the desktop and
program window. Microsoft should be doing the same with speech.

Why? There's almost no demand for it.

Speech recognition is the difficult part. Screen reading is not
nearly as difficult.

Both are extremely difficult if you want truly integrated solutions.

That sounds like pure guesswork and it's beside the point.

It's not pure guesswork. Virtually without exception, putting in
features for a tiny minority of users is a net loss. Companies only
do it for PR, out of corporate conscience, or when the law requires
it. They certainly don't do it to make money.

I guess it depends on whether you believe the disabled people
should have equal access.

Within reason, I believe they should. But I do not believe that vast
resources should be spent on accommodating them when the same
resources could do more good for a larger number of people if spent in
a different way.

Judging solely by your own experience I'm sure. Speech input is way
superior here on my machine. I hate to say this, and that's what I'm
doing, but current technology does require a good speaking voice and
the ability to properly configure sound input.

That's the easy part. Just as generating sound is the easy part of
speech synthesis. The hard part is compressing information into an
audio channel, and making sense of input or reformatting output to fit
it.

Sorry, but you're just making excuses for your own inability to use
the current technology.

Which technology am I unable to use?

There is no such special hardware. There are no special drivers.

If you want to do it right, you need hardware solutions.

Microsoft can bully its way into anything that has to do with
personal computers.

No, it can't. There are a lot of clever and/or well-funded
competitors out there. Not every company is as stupid as Netscape.

There isn't any money in producing software that Microsoft can
integrate into Windows. I guess Microsoft is limited to the most
popular software in order to be less obvious about it.

Microsoft builds what sells. That's business.

But in fact, other software publishers have already produced better
speech software that Microsoft, even though speech is a valid part
of the operating system.

Speech is no more a "valid" part of the operating system than text.

And indeed, whether Microsoft sees money in it is the question.

Whether any company does. For extreme niche markets, small companies
are usually better at turning a profit than large companies.

The future is not a charity.

The future will be just like the present.

In a prior post, you suggested the author you were replying to
should start writing applications for a different operating system.
Of course that's impractical because he wouldn't sell any copies. In
this argument, you are very much aware of the fact that costs of
development must be recovered.

In that prior post, I was making it obvious why people _don't_ write
applications for obscure operating systems.

--
Transpose mxsmanic and gmail to reach me by e-mail.