Does WSR have a speed versus accuracy control like Dragon does?

I can't seem to find the speed vs accuracy control
on Vista Speech. (WSR)

Does such a control exist?

If so, is the accuracy affected if I adjust it for
maximum speed?

Mark-

 

"Mark Conrad" wrote:
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> I can't seem to find the speed vs accuracy control
> on Vista Speech. (WSR)
>
> Does such a control exist?
>
> If so, is the accuracy affected if I adjust it for
> maximum speed?<!--colorc--><!--/colorc-->

Not necessary for WSR as it has very low resource overhead, under 87
megabytes on the computer I'm using now. Dragon uses almost 4 times more.

Marty

 

In article <A4029828-54A7-4EFA-AF62-339B1E115423@microsoft.com>,
<wwwMyMSSpeechcom@discussions.microsoft.com> wrote:
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Not necessary for WSR ...<!--colorc--><!--/colorc-->

Thanks Marty, obviously I got WSR confused with Dragon.

Okay, thanks to you, I now know that WSR does not have
the speed versus accuracy adjustment feature.

I switch between 3 different speech constantly here,
so confusion results sometimes.

The 3 apps are:

1) WSR
2) Dragon
3) MacSpeech

Both Dragon and MacSpeech have a speed versus accuracy
adjustment.

In my limited experimentation, there was no observable
increase in accuracy when I moved the slider all the way
over to the "accuracy" extreme.

As a result of my admittedly crude experiments, I now
leave the slider adjustment all the way over to
the maximum speed position.


Marty, as a point of information, I feel I should elaborate
somewhat more, so kinda forgive my long-windedness.

This was my test sentence:
"now is the time for all good men to come to the aide of their country"

In other words, I was trying to force all 3 speech apps to
output "aide" instead of "aid".

Despite repeated and prolonged training, using the correction
window of *ALL* the 3 speech apps, I had no success.
(hundreds of corrections)

Before someone pops in and says the test sentence is
illegitimate, let me point out that the test sentence
would be legitimate if used in the following context:

The phrase "now is the time for all good men to come
to the aide of their country" is an example of
improper grammar.


No success, by using the correction window _only_,
no macros, no advanced Dragon features such as
"written form" vs "spoken form"...

Correction window only, repeatedly correcting
over and over and over...

I was only able to get the apps to _sometimes_ output
the word "aide", however no permanent results occurred,
especially if I restarted the speech app.

And yes, I saved my supposedly trained profile often.

And yes, I tried many different variations of "correcting",
such as spell mode, typing correction in, etc., etc.



So as a result I gave up, concluded that there was no
possible reason for me to set the slider to the
extreme "accuracy" setting.


As regards the effect of the accuracy setting to display
text onscreen faster, when set to the "fastest" setting,
I detected no marked difference.

I can easily overwhelm all 3 speech apps with rapid
speaking, at say 288 wpm.

(18 repetitions of the test sentence
when I used "aid" in the test sentence)

18 x 16 = 288 wpm

(with practice, most people, myself included, can
dictate at 384 wpm, for 5 seconds)

....but we are considering dictating for a full 60 seconds
------------------------------------
All 3 speech apps refused to display _any_ text
onscreen for one full minute, after which all
288 words popped onscreen

BTW, there was not even one text mistake it the
resulting 288 words, which says a lot
for my enunciation.

Of course, if you take too long to gulp a breath
every two repitions, THEN the text will jump
onto the screen in all 3 speech apps, in response
to the delay caused by your breathing.

Mark-

 

In article <170920092118293101%mconrad@earthlink.invalid>, Mark Conrad
<mconrad@earthlink.invalid> wrote:
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Of course, if you take too long to gulp a breath
> every two repitions...<!--colorc--><!--/colorc-->

Good Grief, I had better correct that, S/B "repetitions",
otherwise the stupid spelling police will get on my case.

My usenet reader does not have a spell checker.


Just for fun...
********
Using the $1,600 Dragon NaturallySpeaking app,
I dictated the name "George Stephanopoulos"
over and over again, 180 repetitions of that name
which took me exactly 120 seconds to dictate.

90 wpm for two minutes - - - allowing for the fact
that "George Stephanopoulos" has 3 times more
syllables than a simple name like "Bob Jones",
that would be a dictation speed of 270 wpm!

*NOTHING* occurred onscreen, until the very end
of the 120 seconds it took me to dictate.

Two full minutes, nothing onscreen!

At the very end of the two minutes, all 180 repetitions
of the name "jumped" onscreen.
(no errors, BTW)

I had the speed/accuracy slider set all the way to
maximum speed.



Repeated the same test, this time with the
"MacSpeech Dictate 1.5.2" medical speech app,
cost $600, same results.
(speed/accuracy slider set for maximum speed)


Repeated the same test yet again with WSR (Vista speech)
but this time I only dictated the name for 60 seconds,
to allow for the fact that WSR is just a consumer app,
so I did not want to risk overloading its speech buffer.

Speed of dictation still 90 wpm, same as previous tests.

Same result, nothing appeared onscreen until the very end
when I finished dictating, then all 90 repetitions of the name
"George Stephanopoulos" jumped onscreen at once.

People, don't take my word for this, do the testing yourself.

Just remember, if it takes you longer than a tenth of a second
to gulp a breath, the results will be invalid, because the
speech app will detect the pause and have time to dump the
names onscreen.

Hint- Some people will find it easier to gulp a quick breath
after each repetition of "George Stephanopoulos".

So much for the fable that WSR is faster than Dragon,
when it comes to dumping text onscreen.

Mark-

 

"Mark Conrad" <mconrad@earthlink.invalid> wrote in message
news:180920090820555741%mconrad@earthlink.invalid...<!--coloro:blue--><span style="color:blue <!--/coloro-->
> In article <170920092118293101%mconrad@earthlink.invalid>, Mark Conrad
> <mconrad@earthlink.invalid> wrote:
> that would be a dictation speed of 270 wpm!
>
>
> Repeated the same test yet again with WSR (Vista speech)
> but this time I only dictated the name for 60 seconds,
> to allow for the fact that WSR is just a consumer app,
> so I did not want to risk overloading its speech buffer.
>
> Speed of dictation still 90 wpm, same as previous tests.
>
> Same result, nothing appeared onscreen until the very end
> when I finished dictating, then all 90 repetitions of the name
> "George Stephanopoulos" jumped onscreen at once.
>
> People, don't take my word for this, do the testing yourself.<!--colorc--><!--/colorc-->

I have, ages ago. When I use WSR, I can get at least twice that and I would
not call myself a fast speaker because, I think before I speak as to what I
want to say and that's the truth. My speed from when I first started using
WSR until now, has quadrupled. Of course, I am now using the very best
desktop Mic. you can buy and that helps. AND with the right tools and
training, it is very accurate. I get 99% accuracy now, no problem.

I'm very satisfied with it and prefer it to Dragon. Windows 7 will be out
in October and I cannot wait to get a copy. Some great reviews in there
regarding the improvements with SR.
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> So much for the fable that WSR is faster than Dragon,
> when it comes to dumping text onscreen.<!--colorc--><!--/colorc-->

That fable is in your head!

Consistantly, grammer, Vitrolic, mistakingly, Dispite, embarassed,
repititions, preceeding.- Spelling mistakes - All dictated by Dragon
Naturally Speaking Medical.

I noticed he did not mention that! What a waste of money.

So much for the fable that Dragon Naturally Speaking Medical can spell all
its words properly, when dictated to the screen.

andy t

 

In article <eqP#UsrOKHA.4028@TK2MSFTNGP05.phx.gbl>, andy t
<speech@dragon_can't_spell_properly.invalid> wrote:
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
> > Repeated the same test yet again with WSR (Vista speech)
> > but this time I only dictated the name for 60 seconds,
> > to allow for the fact that WSR is just a consumer app,
> > so I did not want to risk overloading its speech buffer.
> >
> > Speed of dictation still 90 wpm, same as previous tests.
> >
> > Same result, nothing appeared onscreen until the very end
> > when I finished dictating, then all 90 repetitions of the name
> > "George Stephanopoulos" jumped onscreen at once.
> >
> > People, don't take my word for this, do the testing yourself.<!--colorc--><!--/colorc-->
>
> I have, ages ago. When I use WSR, I can get at least twice that
> and I would not call myself a fast speaker...<!--colorc--><!--/colorc-->

Wow! - at least twice the speed of 90 wpm when speaking the
test name "George Stephanopoulos", that would be 180 wpm.

Hard for me to believe that; in fact, hard for anyone else to
believe that either, if they tried to speak 180 repetitions of
"George Stephanopoulos" in 60 seconds!!!



<!--coloro:blue--><span style="color:blue <!--/coloro-->
> I think before I speak as to what I
> want to say and that's the truth.<!--colorc--><!--/colorc-->

In that case your sense of logic must be really screwed up,
else you would not have just now posted a disconnected
sentence like the following:<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Consistantly, grammer, Vitrolic, mistakingly, Dispite, embarassed,
> repititions, preceeding.- Spelling mistakes - All dictated by Dragon
> Naturally Speaking Medical.<!--colorc--><!--/colorc-->
Shows very flawed "thinking" on your part.

Least you should have done is to identify the context,
if any, of your babbling.


<!--coloro:blue--><span style="color:blue <!--/coloro-->
> My speed from when I first started using
> WSR until now, has quadrupled.<!--colorc--><!--/colorc-->

Now that I can believe, with experience comes speed,
up to the point where WSR limits your speed with its
slow and awkward text correction features.


<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Of course, I am now using the very best
> desktop Mic. you can buy and that helps.<!--colorc--><!--/colorc-->

Tsk, tsk, multiple errors there.
1) improper capitalization of mic'
2) no apostrophe at the end of mic'
3) improper use of a period after mic'
in the middle of a sentence

Good thing you have me looking over your shoulder
in order to correct your flawed posting.

<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Of course, I am now using the very best
> desktop Mic. you can buy and that helps.<!--colorc--><!--/colorc-->

Glad you have that emotional attachment to your mic'.

Being more realistic than you, I realize "the very best"
mic' is a subjective issue which depends entirely on
what the mic' is being used for, and the conditions
under which that mic' is being used.

For speech recognition, my only concern is whether
the mic' I use has any effect on my average text accuracy,
under specified test conditions.

If two different microphones show no detectable difference
in average text accuracy, then as far as I am concerned
they are equal to each other, as far as text accuracy
is concerned.

<!--coloro:blue--><span style="color:blue <!--/coloro-->
> I get 99% accuracy now, no problem.<!--colorc--><!--/colorc-->

Same here, in general terms, so I am happy.

My accuracy varies wildly, depending on many factors,
too numerous to list.

<!--coloro:blue--><span style="color:blue <!--/coloro-->
> I'm very satisfied with it and prefer it to Dragon.<!--colorc--><!--/colorc-->

I prefer Dragon, which has many more text correction
features which speed up the correction of the inevitable
text mistakes, which in turn speeds up the overall speed
that text can be produced from speech.

Take away that speed benefit, then the average user has
little incentive to use speech recognition at all.


<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Windows 7 will be out
> in October and I cannot wait to get a copy.<!--colorc--><!--/colorc-->

Same here, I run Windows on Mac hardware, of course.

<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Some great reviews in there
> regarding the improvements with SR.<!--colorc--><!--/colorc-->

I hear mixed reviews, some experts such as "Chucker"
in the Knowbrainer forums claim there is little or no
difference, unless a user switches to an "all 64 bit OS".

Myself, I would welcome switching to 64 bits, however
think I will wait until more drivers are available
for 64 bit apps.

I do not know if/when WSR will be recoded to take full
advantage of 64 bit speed.

Nuance has come right out and stated that they have no plans
to recode Dragon for 64 bit operation until the distant future,
when 64 bit operation becomes more common.

<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
> > So much for the fable that WSR is faster than Dragon,
> > when it comes to dumping text onscreen.<!--colorc--><!--/colorc-->
>
> That fable is in your head!<!--colorc--><!--/colorc-->

Afraid to put it to the test eh - - - figures.


Would be easy enough for you to test.

All that is required is for you to produce a
long single "utterance" without any detectable
dead space where you are grabbing a breath.

Remember, any pause greater than about a fifth
of a second, will cause _all_ _three_
speech applications to immediately dump text
onto the screen, WSR, Dragon, and MacSpeech.

(at least it does using my 2-yr old MacBook Pro,
no difference at all between the speech apps)



I know that I can _easily_ speak for 5 seconds
without needing to gulp a breath, so likely you can
also?

If so, all you need to do is to borrow a friends
digital recorder.

Well here we might have a hitch regarding your
personality, you probably have no friends.

In the unlikely event you have a friend with
a recorder he will let you borrow:

Record 12 segments of speech, being certain to
turn on the record switch the instant you start
a 5 second recording session - - - then turn off
the record switch before you need to breathe.

If you do it right, you will wind up with 60 seconds
of continuous speech, with no detectable pauses
between any of the 5 second "sessions".

Play the resulting audio directly into the mic'
of your WSR speech app, I just did so here.

WSR *DOES* refuse to display text for 60 seconds,
same as Dragon, same as MacSpeech,
*NO* difference at all.

It was downright amazing how accurate the text from
_my_ digital recording was, when WSR _finally_
converted the continuous 60 second recording to text
and dumped that text into my open NotePad document.



When you get the results of this test I might accept
your apology for your asinine statement in this thread.


BTW, here I am not advocating using a digital recorder
to produce text in the fashion above, as there are many
more "accurate" ways of getting a digital recorder to
produce text.

Using the best possible hardware and software, it is
possible to get average raw accuracy of about 98%,
which is about 1% _worse_ than dictating directly
into a speech app' by conventional means.

Mark-

 

In article <eqP#UsrOKHA.4028@TK2MSFTNGP05.phx.gbl>, andy t
<speech@dragon_can't_spell_properly.invalid> wrote:
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> So much for the fable that Dragon Naturally Speaking
> Medical can spell all its words properly,
> when dictated to the screen.<!--colorc--><!--/colorc-->

You have been told repeatedly by me that I _never_ use
Dragon to write these usenet posts.

You also know, or should know, that Dragon Medical is used
for medical record keeping in the USA.

I know of not even one hospital in the US that uses WSR,
out of the thousands of hospitals here.


About the cost of Dragon Medical at $1,600 USD
---------------------------<!--coloro:blue--><span style="color:blue <!--/coloro-->
> What a waste of money.<!--colorc--><!--/colorc-->

Hospitals and individual doctors do not think so,
which is why Nuance is making hundreds of millions
of dollars profit each year, mainly from their medical
speech recognition software.

Now even MacSpeech is getting into the lucrative medical
market, with their $600 medical version for the Mac.
(which I have, BTW)

Medical side industries are springing up also, such as
"MacPractice MD" for the Mac. They sell a minimum
software app for $2,000 but by the time auxiliary apps
are added, the customer has spent $5,000.

All "MacPractice MD" does is to assist medical users in
filling out EMR and EHR forms.

The MacPractice company advertises in their website that
"MacPractice MD" works well with "MacSpeech".

Last time I checked the MacPractice company was making
money hand over fist with their medical software, and they
have only offered it for the recent few years.

So much for your wrong statement that medical software
is a waste of money.


When it is my turn to get medical care, I do not want to
wait two days while some smelly guy wearing a turban and
a big diaper is processing my medical records.

Representative transcription site, showing how screwed up
the time-wasting situation is with offshore transcription:

<http://transcribe.inteleants.com/?gclid=CO3euK2j_pwCFVhh2godXlZrbg>


Not to mention the 200,000 medical mistakes that occur
each year in the USA. (UK only has about 100,000)

A lot of those mistakes occur because doctors forget to
document medical info' promptly, instead they try to
remember it later.

....for example, from some overworked ER doctor who noticed
that his 8am emergency bleeder patient was an Indian national,
but forgot to recall that fact late at night when he was trying to
remember the events of the morning for his medical reports.

....so about a week later the Indian patient died because there were
no records that he had a possible "Bombay" blood type, which only
about a thousand people in the entire world have...



....a fact that our overworked ER doctor should have promptly
documented, instead of relying on his memory hours later
when he was doing his typing late at night.

Usual procedure is to dump O positive blood into a patient,
which will work in all common cases, _except_ for rare cases
like the Bombay blood type.

O positive will save his life in an emergency, buy as soon as
possible his blood has to be drained and replaced with blood
from "Bombay type" donors, who are _very_ hard to find
given there are only a thousand worldwide.

Blood would have to be flown in, from one of the extremely few
blood banks that have access to that blood type.

Mark-

 

"Mark Conrad" <mconrad@earthlink.invalid> wrote in message
news:210920091754085078%mconrad@earthlink.invalid...<!--coloro:blue--><span style="color:blue <!--/coloro-->
> In article <eqP#UsrOKHA.4028@TK2MSFTNGP05.phx.gbl>, andy t
> <speech@dragon_can't_spell_properly.invalid> wrote:<!--colorc--><!--/colorc-->

<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro--><!--coloro:darkred--><span style="color:darkred <!--/coloro-->
>> > Repeated the same test yet again with WSR (Vista speech)
>> > but this time I only dictated the name for 60 seconds,
>> > to allow for the fact that WSR is just a consumer app,
>> > so I did not want to risk overloading its speech buffer.
>> >
>> > Speed of dictation still 90 wpm, same as previous tests.
>> >
>> > Same result, nothing appeared onscreen until the very end
>> > when I finished dictating, then all 90 repetitions of the name
>> > "George Stephanopoulos" jumped onscreen at once.
>> >
>> > People, don't take my word for this, do the testing yourself.<!--colorc--><!--/colorc-->
>>
>> I have, ages ago. When I use WSR, I can get at least twice that
>> and I would not call myself a fast speaker...<!--colorc--><!--/colorc-->
>
> Wow! - at least twice the speed of 90 wpm when speaking the
> test name "George Stephanopoulos", that would be 180 wpm.
>
> Hard for me to believe that; in fact, hard for anyone else to
> believe that either, if they tried to speak 180 repetitions of
> "George Stephanopoulos" in 60 seconds!!!<!--colorc--><!--/colorc-->

The same reason is hard for me to believe your Dragon tests. You cannot
prove them or disprove my claims.

Believe it, I could go faster if I wanted to using WSR, but I do not dictate
that way and nor do many other people except you, but then you are an
exception to the human race!
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
>> I think before I speak as to what I
>> want to say and that's the truth.<!--colorc--><!--/colorc-->
>
> In that case your sense of logic must be really screwed up,
> else you would not have just posted a disconnected
> sentence like the following:<!--colorc--><!--/colorc-->

It was not meant to be a sentence but a list, you senile old illiterate.
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
>> Consistantly, grammer, Vitrolic, mistakingly, Dispite, embarassed,
>> repititions, preceeding.- Spelling mistakes - All dictated by Dragon
>> Naturally Speaking Medical.<!--colorc--><!--/colorc--><!--colorc--><!--/colorc-->
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Shows very flawed "thinking" on your part.
>
> Least you should have done is to identify the context,
> if any, of your babbling.<!--colorc--><!--/colorc-->

I believe I got my point across and it did its job.
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
>> My speed from when I first started using
>> WSR until now, has quadrupled.<!--colorc--><!--/colorc-->
>
> Now that I can believe, with experience comes speed,
> up to the point where WSR limits your speed with its
> slow and awkward text correction features.<!--colorc--><!--/colorc-->

To be honest, I do not care what you think or believe. It is fast enough
for me, faster than typing and I prefer it to Dragon. Get over it.
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
>> Of course, I am now using the very best
>> desktop Mic. you can buy and that helps.<!--colorc--><!--/colorc--><!--colorc--><!--/colorc-->
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Tsk, tsk, multiple errors there.
> 1) improper capitalization of mic'
> 2) no apostrophe at the end of mic'
> 3) improper use of a period after mic'
> in the middle of a sentence<!--colorc--><!--/colorc-->

Lol! Is that the best you can do?
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Good thing you have me looking over your shoulder
> in order to correct your flawed posting.<!--colorc--><!--/colorc-->

Yeah! He who professes to know the meaning of déjà vu and whose spelling
using Dragon is atrocious. I'm so glad you're looking over my shoulder.
It's a world of comfort. Lol!
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
>> Of course, I am now using the very best
>> desktop Mic. you can buy and that helps.<!--colorc--><!--/colorc-->
>
> Glad you have that emotional attachment to your mic'.<!--colorc--><!--/colorc-->

It is not an emotional attachment; it is just a statement of fact.
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Being more realistic than you, I realize "the very best"
> mic' is a subjective issue which depends entirely on
> what the mic' is being used for, and the conditions
> under which that mic' is being used.<!--colorc--><!--/colorc-->

It is not that difficult to understand really but I can see for someone like
you, that everything is difficult to understand.

The Sennheiser 431 ll microphone is designed specifically for speech
recognition. Fact. It is also known worldwide to be the best desktop
microphone on the market. Fact.






<!--coloro:blue--><span style="color:blue <!--/coloro-->
> For speech recognition, my only concern is whether
> the mic' I use has any effect on my average text accuracy,
> under specified test conditions.
>
> If two different microphones show no detectable difference
> in average text accuracy, then as far as I am concerned
> they are equal to each other, as far as text accuracy
> is concerned.<!--colorc--><!--/colorc-->

I have three microphones. All of them have their own characteristic
advantages.

The Sennheiser ME3 headset microphone is extremely accurate and noise
cancelling, mainly because the microphone element is right by the mouth and
therefore cancels out any noise around it. I would say it is on a par with
the MD 431 ll. The only problem for me is that I find headset microphones
uncomfortable to wear and the fact that I cannot take them off physically
because of my disability.





The Buddy Desktop microphone is also very accurate but not so noise
cancelling. It is very prone to picking up noises around it. Apart from
that, it has the advantage of sitting on a desktop and not taking up much
room.

So you see, there are differences with different microphones and they DO
make a difference to the accuracy of speech recognition, no matter what you
say or what gibberish you spill.
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> I hear mixed reviews, some experts such as "Chucker"
> in the Knowbrainer forums claim there is little or no
> difference, unless a user switches to an "all 64 bit OS".<!--colorc--><!--/colorc-->

We beg to differ then.
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Myself, I would welcome switching to 64 bits, however
> think I will wait until more drivers are available
> for 64 bit apps.<!--colorc--><!--/colorc-->

I do not know enough about 64 bit apps or what it means except to say, that
a 64 bit OS is better for gaming or stuff like that.
Perhaps you could enlighten me on what the advantages of a 64 bit app can
do?

On second thoughts, don't bother. I will probably get a 16 page essay full
of nonsense as usual!
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro--><!--coloro:darkred--><span style="color:darkred <!--/coloro-->
>> > So much for the fable that WSR is faster than Dragon,
>> > when it comes to dumping text onscreen.<!--colorc--><!--/colorc-->
>>
>> That fable is in your head!<!--colorc--><!--/colorc--><!--colorc--><!--/colorc-->
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Afraid to put it to the test eh - - - figures.<!--colorc--><!--/colorc-->

Oh please! Isn't it time for you to grow up? You are an 80 odd year old
man for Christ's sake. Start acting it!

I do not have to and will not pander to your childish tests. How old are
you in the head, 3?

I feel quite sorry for you really. You are quite sad and you must be
lonely. Still, look on the bright side; you still have death to look
forward to!

andy t

 

"Mark Conrad" <mconrad@earthlink.invalid> wrote in message
news:210920092256252762%mconrad@earthlink.invalid...<!--coloro:blue--><span style="color:blue <!--/coloro-->
> In article <eqP#UsrOKHA.4028@TK2MSFTNGP05.phx.gbl>, andy t
> <speech@dragon_can't_spell_properly.invalid> wrote:<!--colorc--><!--/colorc-->
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
>> So much for the fable that Dragon Naturally Speaking
>> Medical can spell all its words properly,
>> when dictated to the screen.<!--colorc--><!--/colorc-->
>
> You have been told repeatedly by me that I _never_ use
> Dragon to write these usenet posts.<!--colorc--><!--/colorc-->

Prove it!

You can't. The same way you cannot prove your Dragon speed tests. We have
only your word that it is true and I for one do not believe you!

< SNIP >

BLAH! BLAH! BLAH!

Good grief! Same OT drivel! For your own sanity, go and get some help, you
silly old man!

I notice that when you post in here now, nobody replies anymore because
nobody cares. You are still talking to yourself in a Microsoft forum. Good
luck with the treatment!

Good luck to you and your imaginary friends! Give my regards to Harvey!
LOL!

andy t

 

In article <Od1AkM6OKHA.1280@TK2MSFTNGP04.phx.gbl>, andy t
<speech@dragon_can't_spell_properly.invalid> wrote:
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
> > Hard for me to believe that; in fact, hard for anyone else to
> > believe that either, if they tried to speak 180 repetitions of
> > "George Stephanopoulos" in 60 seconds!!!<!--colorc--><!--/colorc-->
>
> The same reason is hard for me to believe your Dragon tests.
> You cannot prove them or disprove my claims.<!--colorc--><!--/colorc-->

Ah I see, so you tactically admit that your 180 repetitions
of speaking "George Stephanopoulos" in 60 seconds is a lie.

Best I could do when I tried was 132 repetitions, however
if _those_ were spoken into WSR the accuracy would be
impossibly poor, besides having no bearing on your wrong
claim that WSR can display text in a document _quicker_
than Dragon can, in a practical dictation session.

One would think that you would give up on that outrageous
claim when it is so easy for _anyone_ to see with their
own eyes that a five-second continuous utterance never
appears in a document *UNTIL* a user pauses
for a breath.

Same with a 10 second utterance, or a 30 second, or a
60 second, or until the WSR buffer blows up (overflows)
with a longer continuous utterance.

All 90 of my repetitions in 60 seconds were error free.

....which is _easy_ for any WSR expert to "prove or disprove",
even you, without resorting to lies as you do.

Should be dead easy for you to verify my 90 repetition claim,
being you already posted that you can speak 180 repetitions
in 60 seconds.

....unless you were lying about your 180 claim.


If you had even the sense of a rock, you would realize how
futile it is to attempt to do two brain-consuming tasks
at the same time, such as concentrating on what you plan
to say next, while being distracted by attempting to correct
the onscreen text.

But you do not have the sense of a rock, which is obvious.

Correcting text is a consuming full time brain intensive
activity, especially when peoples lives hinge on the
accuracy of text correction.

It can not be done "on the fly" while one is dictating,
without risking peoples lives.

That is one important reason why WSR will forever be
a fringe app for careless hobbyists, instead of a serious
business speech application like Dragon.

WSR lacks many important text correction features,
such as the ability to play back the original audio which
is synchronized with the words as the audio and words
are "played back" for text correction purposes.


Andrew carelessly posted:
-----------------
Consistantly, grammer, Vitrolic, mistakingly, Dispite, embarassed,
repititions, preceeding.- Spelling mistakes - All dictated by Dragon
Naturally Speaking Medical.
-----------------

Then promptly lied by claiming it was not a sentence,
in this remark:
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> It was not meant to be a sentence but a list, you senile old illiterate.<!--colorc--><!--/colorc-->


Tsk, tsk, don't burst a blood vessel Andrew, getting so upset when
I correct you, as you constantly need my oversight to
keep you ontrack.

I know you are dumb as a rock, so I will make it easier
for you to comprehend the difference between
a sentence and a list:

sentence (noun)
-----------
A set of words that is complete in itself, typically
containing a subject and predicate, question, exclamation,
or command, and consisting of a main clause and sometimes
one or more subordinate clauses.


list (noun)
-------
A number of connected items or names written or printed
consecutively, typically one below the other.


Now I know it will be a real stretch for your limited
intelligence, but try to comprehend the difference, Andrew.

<!--coloro:blue--><span style="color:blue <!--/coloro-->
> To be honest, I do not care what you think or believe.<!--colorc--><!--/colorc-->

Yes you do, else you would not devote so much of your
time and effort to these threads.


<!--coloro:blue--><span style="color:blue <!--/coloro-->
> It is fast enough for me, faster than typing
> and I prefer it to Dragon. Get over it.<!--colorc--><!--/colorc-->

Get over What?

It does not bother me that you prefer WSR.

I use WSR myself, also 3 different versions of Dragon,
and 2 different versions of MacSpeech Dictate.

<!--coloro:blue--><span style="color:blue <!--/coloro-->
> I'm so glad you're looking over my shoulder.
> It's a world of comfort.<!--colorc--><!--/colorc-->

That's the spirit, Andrew, glad you came to your senses.


<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro--><!--coloro:darkred--><span style="color:darkred <!--/coloro-->
> >> Of course, I am now using the very best
> >> desktop Mic. you can buy and that helps.<!--colorc--><!--/colorc-->
> >
> > Glad you have that emotional attachment to your mic'.<!--colorc--><!--/colorc-->
>
> It is not an emotional attachment; it is just a statement of fact.<!--colorc--><!--/colorc-->

Andrew, Andrew, Andrew, when will you ever learn!
(sheesh, Andrew is dumb as a rock)
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> it is just a statement of fact.<!--colorc--><!--/colorc-->
No it is not a statement of fact, as you can prove to
yourself when you try to use that "very best desktop mic' "
in a very noisy environment, such as a bunch of loud kids
running around in your room, or a very loud TV in your room.

I can dictate fast accurate text in such a noisy environment,
but not by using the desktop mic' that you use.

I can use yet another different desktop mic' at a distance of 50
inches from my mouth in a dead quiet environment, but not by
using the mic' that you use.

Making a blanket statement that any mic' is
"the very best desktop mic' " only reinforces my opinion
that you are as dumb as a rock.

Rational people realize there is no "very best desktop mic' "
as it depends on the exact conditions in place
when one uses their desktop mic'.

Change the conditions, and the "very best desktop mic' " often
falls flat on its face.


<!--coloro:blue--><span style="color:blue <!--/coloro-->
> The Sennheiser ME3 headset microphone is
> extremely accurate...<!--colorc--><!--/colorc-->

You are switching topics, because all the discussion above
is your asinine statement "the very best *DESKTOP* mic' "

Mark-

 

On 9/22/09, the entity Mark Conrad wrote this:<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Andrew carelessly posted:
> -----------------
> Consistantly, grammer, Vitrolic, mistakingly, Dispite, embarassed,
> repititions, preceeding.- Spelling mistakes - All dictated by Dragon
> Naturally Speaking Medical.
> -----------------<!--colorc--><!--/colorc-->
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Then promptly lied by claiming it was not a sentence,
> in this remark:<!--colorc--><!--/colorc-->
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
>> It was not meant to be a sentence but a list, you senile old illiterate.<!--colorc--><!--/colorc--><!--colorc--><!--/colorc-->

<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Tsk, tsk, don't burst a blood vessel Andrew, getting so upset when
> I correct you, as you constantly need my oversight to
> keep you ontrack.<!--colorc--><!--/colorc-->

Well, I had no difficulty in seeing that it was a list and not a
sentence when I read it...

It was rather obviously a list of errors that the program had
generated.

--
Gene Bloch 650.366.4267 lettersatblochg.com

 

In article <#6aUuO6OKHA.508@TK2MSFTNGP06.phx.gbl>, andy t
<speech@dragon_can't_spell_properly.invalid> wrote:
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
> > You have been told repeatedly by me that I _never_ use
> > Dragon to write these usenet posts.<!--colorc--><!--/colorc-->
>
> Prove it!<!--colorc--><!--/colorc-->

That is easy. Your own malformed sentence:

Andrew carelessly posted:
-----------------
Consistantly, grammer, Vitrolic, mistakingly, Dispite, embarassed,
repititions, preceeding.- Spelling mistakes - All dictated by Dragon
Naturally Speaking Medical.
-----------------


*ALL* those words in your malformed sentence can be looked up
by anyone who has Dragon, and *ALL* those words will be
missing from the Dragon vocabulary.

It is therefore impossible for Dragon to generate "grammer"
for example because "grammer" is *NOT* in its vocabulary.

Dragon will generate the correctly spelled "grammar" instead.

Same for all the other examples in your malformed sentence.


<!--coloro:blue--><span style="color:blue <!--/coloro-->
> Prove it!
> You can't.<!--colorc--><!--/colorc-->

I just did.



<!--coloro:blue--><span style="color:blue <!--/coloro-->
> You can't. The same way you cannot prove your Dragon speed tests. We
> have
> only your word that it is true and I for one do not believe you!
>
> < SNIP ><!--colorc--><!--/colorc-->

Again you are sloppy in your posting, the same as you were when
you confused a sentence with a list.
(comes from you being as dumb as a rock)

You failed to specify which "speed tests" you were referring to.

All of them?
One particular test?

Being you failed to specify, I can only guess that you mean
you do not believe anything I post about my experience with
"Dragon speed tests".

That is your choice, but it is a rather stupid choice.
(comes from you being as dumb as a rock)

I have no reason for lying about such matters, because I am
just as quick to point out Dragons flaws
as anyone else here is.

Dragon has plenty of flaws, despite its high price and
overwhelming acceptance by the medical and other
professional users.

As regards speed of dictation, it is my experience that
all 3 major speech apps get the text into a document in
about the same amount of time, which fact can be "proved"
by anyone who efficiently uses all 3 major modern apps,
Dragon, WSR, and MacSpeech.

I run all 3, very well, I might add.

Still posting here about *SPEED*.

Where both WSR and MacSpeech fall short in "speed" is
their lack of correction features, once the raw text
is ready to be corrected.

It does little good to dictate at 160wpm if the correction
process drags the overall speed down to 40wpm.

Might as well type it in and save all the money you
would have wasted on speech recognition.

Again, my experience, YMMV.

For example, a doctor can easily make all his rounds
wearing a wireless mic' like the "Airline 77" or its
newer counterpart in the 642-645MHz band.
(less subject to being swamped by
any nearby digital TV station which
might afflict the "regular" Airline 77 mic'
which operates in the same 800MHz range)

The doctors critical examination regarding my prior
post about an Indian national "bleeder" patient in the
emergency room would get immediately transmitted
to a medical "correctionist" whose prime job is to
handle and correct the dictation from the ER doctor
in real time.

Medical tests would be immediately ordered to make
_certain_ the Indian "bleeder" was not a rare
"Bombay phenotype" person, thereby saving his life
after an immediate transfusion of the wrong blood type
"O positive" was given in order to save his life,
until such time as the rare "Bombay phenotype" of blood
can be flown in for the correct permanent transfusion.

Bottom line -
-------
In effect our emergency room doctor can transmit
dictation all shift long at his normal speaking rate
of 100-160wpm, with occassional bursts at
200wpm or faster, such as the hectic cases of
cardiac arrest.

The "correctionist" in a nearby room acts like a
traffic cop, correcting the doctors text in real time
as the doctor speaks, routing critical corrected text
to nearby hospitals, labs, medical consultants,
or in the case of our Indian "bleeder" arranging for
a complex blood test for a rare blood type, then
arranging to fly that rare blood from thousands
of miles away in time to save the Indian from death.

All the above can not be done with any speech app
except Dragon, for many obvious reasons.


For poorer medical institutions, the ER doctor
could dictate directly into a dedicated digital
recorder, such as the Olympus DS-5000
which I own, widely recognized as the best
digital recorder for speech recognition,
replacing the venerable Philips 9600.

Doctor would have to periodically swap the
memory card to his "correctionist", in order
to approximate real time correction of his text,
such as in the case of the Indian "bleeder",
or a snake bite victim, or any other case where
almost instant processing of his medical text
is required.

Bottom bottom line -
------------
All the above would be almost impossible
to do with WSR, Dragon would be required.

Speed is essential for normal uses of speech
recognition, excepting a few niche uses of
speech recognition where average speed
over 45wpm is of no concern.

Lacking speed, a person is better off just using
regular typing into a word processor.

Mark-

 

"Gene E. Bloch" <letters@someplace.invalid> wrote in message
news:#Oqck79OKHA.5488@TK2MSFTNGP02.phx.gbl...
<!--coloro:blue--><span style="color:blue <!--/coloro-->
> On 9/22/09, the entity Mark Conrad wrote this:<!--colorc--><!--/colorc-->
<!--coloro:blue--><span style="color:blue <!--/coloro--><!--coloro:green--><span style="color:green <!--/coloro-->
>> Andrew carelessly posted:
>> -----------------
>> Consistantly, grammer, Vitrolic, mistakingly, Dispite, embarassed,
>> repititions, preceeding.- Spelling mistakes - All dictated by Dragon
>> Naturally Speaking Medical.
>> -----------------<!--colorc--><!--/colorc-->
><!--coloro:green--><span style="color:green <!--/coloro-->
>> Then promptly lied by claiming it was not a sentence,
>> in this remark:<!--colorc--><!--/colorc-->
><!--coloro:green--><span style="color:green <!--/coloro--><!--coloro:darkred--><span style="color:darkred <!--/coloro-->
>>> It was not meant to be a sentence but a list, you senile old illiterate.<!--colorc--><!--/colorc--><!--colorc--><!--/colorc-->
>
><!--coloro:green--><span style="color:green <!--/coloro-->
>> Tsk, tsk, don't burst a blood vessel Andrew, getting so upset when
>> I correct you, as you constantly need my oversight to
>> keep you ontrack.<!--colorc--><!--/colorc-->
>
> Well, I had no difficulty in seeing that it was a list and not a sentence
> when I read it...
>
> It was rather obviously a list of errors that the program had generated.
>
> --
> Gene Bloch 650.366.4267 lettersatblochg.com<!--colorc--><!--/colorc-->

Thank you Gene.

Mark Conrad has difficulty with most things in life and because most people
can see it is patently obvious that it is a list and not a sentence, he will
say black is white just to try and be clever. The problem is, the more he
tries, the worse he becomes.

Most of the things he claims on the screen, is a pack of lies. He proves
nothing!

Quite sad really.

andy t

 

In article <e#guOiCPKHA.508@TK2MSFTNGP06.phx.gbl>, andy t
<speech@dragon_can't_spell_properly.invalid> wrote a confused mess...

Now Andrew, you poor confused useless soul, you
probably do not know the definition of a list.

I better offer to look over your shoulder and guide you
to understand the definition of a list, being you are
still as dumb as a rock.

Mark-

 

"Mark Conrad" <mconrad@earthlink.invalid> wrote in message
news:230920090929265677%mconrad@earthlink.invalid...<!--coloro:blue--><span style="color:blue <!--/coloro-->
> In article <e#guOiCPKHA.508@TK2MSFTNGP06.phx.gbl>, andy t
> <speech@dragon_can't_spell_properly.invalid> wrote a confused mess...
>
> Now Andrew, you poor confused useless soul, you
> probably do not know the definition of a list.
>
> I better offer to look over your shoulder and guide you
> to understand the definition of a list, being you are
> still as dumb as a rock.
>
> Mark-<!--colorc--><!--/colorc-->
|||
You sad old sod. <(°|°)>
(^)
|°|

Say goodnight to the folks Gracie!

andy t