Re: Confession for my nasty abuse of the Dragon speech engine

"Mark Conrad" <> wrote in message
news:251120090648451513%...

> I am ashamed of myself,

There is no need to be, we all make mistakes. Just some more than others.

> ...y'know, where he interminably carries on about
> the really important things in life, such as minor
> spelling errors.

'Minor'? One of your spelling errors was 'dispite'.

That is truly unforgivable from a so-called educated person. Some of the
others were quite shameful but that one takes the bullet.

To be able to take one seriously in any correspondence, spelling and grammar
is very important. I would not call it 'minor' if your spelling was
constantly incorrect. Use of the English language is very important
especially if you own a copy of DNS Legal where an incorrect spelling could
be quite damaging to a case.

However, nobody is going to take anybody to court in these newsgroups for
misspelling a word, two, or three. Nevertheless, spelling is very important
in the grand scheme of things so don't ever say it is minor, which just
proves that you agree that your spelling is atrocious.

What is criminal in all of this is that Dragon is so accurate and cannot
misspell, yet you are still hell bent on typing when you know this can get
you in all sorts of trouble with your atrocious spelling mistakes. This is
the perfect tool for people like you who do not spell that well yet you
refrain from using it. Madness if you ask me.

> Confession is good for the soul,

I agree. I have also got some confessions to make as well.

I finally got the funding for the brand new Dragon Professional 10.1
software and a 'Plantronics Wireless Hands-free Headset CS60-USB' microphone
that just sits over the ear, thrown in for good measure.

I have been learning the commands and trying out the dictation along with
bombarding the KnowBrainer forums with questions. No doubt, you will have
looked and seen that, being the voyeur you are.

It has been just over a week now since I've started using it and come up
with these honest observations.

For accuracy in dictation, I am now willing to concede that Dragon is far
superior in that department. The other function I really love in Dragon is
the fact that you can have other things working in the background such as a
telly or fan and it will not wake up the microphone as it does in WSR. WSR's
sound meter is far too sensitive so the easiest way is to turn WSR off when
dictating long passages of text and just use Dragon for this use.

However, when it comes to the 'command' side of things, WSR is much faster
as to getting around the computer and does not have the delay that Dragon
does.

I would put money on it, especially with the show numbers command in WSR
that I would beat you with hands down in getting around the computer faster.
In that department, Dragon just does not let you free flow from one command
to the other with the delay between each command that it possesses.

WSR has no delay and is far superior in that department. If you read the
posts in the KnowBrainer forums, even Lunis admits that Dragon is trying to
implement their own version of the numbers command in Dragon 11 when it
comes out, but it will not be on the level of the WSR version. This, from a
disabled point of view, is imperative because at no point can I use my hands
to simply type the Key I want if I want to pick a point on the screen at
speed.

I know Dragon does have a numbering system of sorts, but it is nowhere as
efficient as WSR's at the moment.

Another thing that Dragon has problems with that WSR does not and this for
me is a very important function; is that if I am on a webpage and I select
the command 'Select all', it will highlight everything on the web page. Now
with WSR, if I simply open up Microsoft Office Word 2007 and then say
'paste', everything on the web page will come out with no problems at all,
including all the html glorious colour. I can then edit as and when I
please from that document. Not only that, I can do in a matter of seconds.

I have tried this with Dragon and you can try it yourself. Word goes
bananas as soon as you try to edit anything on the page. Word just cannot
take the resource hog that is Dragon. If you read one of the posts in the
KnowBrainer forums, even Lunis admits this.

What you can do in Dragon is say 'Copy to Notepad', then all the writing
without any html code will be placed in notepad, and then you can edit it.
I like to have the option of either.

It is still early days yet, but I have both the Dragon-bar and WSR on at the
same time so I can choose which app I want to use. Now I use Dragon for
dictating and WSR for getting around the screen and doing quick emails as
such. After a while though, the wireless microphone starts to hurt my ear
and I have to revert to a desktop microphone. Over the head or over the ear
can be a disadvantage as I have explained for a disabled person.

> would be a dictation speed of 211 words per minute!

Even what you claim above seems a bit farfetched but it is a little bit more
realistic than your other ridiculous claims.

I have tried your experiment with DNS Pro by reading from a script and
timing my dictation for 1 minute. I did the experiment twice. I read as
fast as I could and at the end of the first experiment, it came out at 142
wpm. There were no spelling mistakes as such but some of the words came out
differently to what I dictated. I believe there were about four or five
mistakes and I put this down to dictating at speed so Dragon could not pick
out the words clearly.

I did the experiment again and spoke as fast as I could once again. This
time it came out at 163 wpm but there were about six or seven words that did
not come out properly once again.

Adding the two together, that averages out at 150 wpm. That is what most
speech experts claim that it can do if you dictate clearly and correctly,
not necessarily at speed where you can easily pronounce a word incorrectly.

The guy that come and put Dragon on my machine and trained me, emphasised
that you must speak clearly and concisely when dictating in Dragon. This is
also the case in WSR or any other speech app.

I challenge anyone to try this. Dictate as fast as you can from a script
using this software and time yourself for 1 minute and I bet you do not get
more than 160-165 wpm if you are lucky. I also bet there are some words
that do not come out properly as you dictate them at speed. Dragon was not
designed to work this way.

> Perfect 100% text, no correction needed !!!

No Speech app at the moment is 100% word perfect, nor do they claim it. I
have told you this before and it is a well-known fact given out by the
speech experts.

You are still living in la la land and I have told you before to stop
daydreaming and telling porky pies!

> Okay "andy t", I am ready to be admonished.

I admire your honesty and ditto.

andy t

"andy t" <> wrote in message
news:#...

Take a look at some of these posts in the KnowBrainer forums. Most people
accept Dragon for dictating, WSR for navigation.

I agree with you now, Dragon is very accurate with solid dictation but it is
nowhere near as fast as WSR for navigation around the computer ESPECIALLY
with the 'Show Numbers' command.

http://www.knowbrainer.com/PubForum/...c&topicId=8160

I told you I was not emotionally attached to just one speech app unlike you.

I have both now so I have the best of both worlds and I am quite happy with
my speech recognition thank you very much ;-)

andy t

"Mark Conrad" <> wrote in message
news:261120091725365181%...

Hi Mark :-)

> Now don't turn all soft and mushy on me, we worked hard to
> develop our present antagonistic relationship.

Just stating facts and what is in front of my eyes . Now I have been given
Dragon Pro and it is brand-new out of the box , I thought why not let me see
what all the fuss is about and so I have made a conscious effort to test out
this software properly.

> ...and as you know, brevity is our hallmark, hee hee.

Don't know what it means :-)

> Oh contrarie, not far fetched, we will have you up to 211wpm
> or higher in jig order, it is just a matter of technique.

What is all this fascination with speed? I am using Dragon now and talking
just at a normal pace and it is working perfectly well and is very accurate.
This is the way it is meant to be used speaking clearly and concisely so
that every word can be recognised. These were the exact words I was told by
the guy tutoring me. I suppose you are now going to say he is wrong and you
are right? I do not need to dictate at 211 wpm.

Adding the fact that I still say it is impossible to dictate that fast and
expect the wording to come out correctly at that speed.

Okay you got me with the accuracy thing. I bow to thee on this one but
reading at speed serves no purpose. At the moment I am just dictating what
comes to my mind. No script, just thinking what to say at a normal pace.
This is perfect for what I need, No need to rush, just making sure that
every word is spoken and recognised by my voice.

> BTW, that perfect 100% accuracy on 1,479 syllables
> in 7 minutes is repeatable; just for kicks I did it 3 times
> in-a-row and still managed to achieve 100% raw accuracy,
> no corrections needed.

And you dictate like that normally do you? Like a rabid dog? I think not.
Nobody does except you.


> (using the flaky MSD speech app, on a 2 yr old MacBook)
>
> I will admit that when I tried it for a 4th time, I messed up
> and badly mispronounced one of those medical terms, causing
> a text mistake.
>
>
>> I did the experiment again and spoke as fast as I could once
>> again. This time it came out at 163 wpm but there were
>> about six or seven words that did not
>> come out properly once again.
>
> All has to do with correction technique, your (prior)
> correction techniques are different from mine,
> so naturally when you make a correction, it tends
> to not necessarily "stick" the next time around.
>
> Also has to do with speaking technique.
>
> I tend to speak staccato, with micro pauses between
> words, for example when I want "it is" to register as
> two separate words, I _never_ speak the one-syllable
> "its" - - - rather I use two syllables: "it tis".

Maybe you're right, but it is working fine just at the moment as I'm
speaking at a normal pace which is still much faster than typing. Answer
this question: If it is working just fine and dandy now and I'm speaking at
normal pace just like I would be if I was talking to somebody, why should I
change to any technique if it is working very well at the moment and it's
very accurate?

Granted, as time goes on and corrections are corrected, it will do the same
as any other speech app and learn from the mistakes thus becoming more
accurate.

>> I challenge anyone to try this. Dictate as fast as you can
>> from a script using this software and time yourself
>> for 1 minute and I bet you do not get more
>> than 160-165 wpm if you are lucky.
>
> Now you have me pinned down.
>
> I will go you one better, I have a hard time dictating
> technical text faster than 86 wpm.

Just test the way that I did it. Read out a letter or something as fast as
you can and time yourself for one minute and then count how many words are
there. Remember, this includes all punctuation. If you can get 211 words per
minute even with your special technique, then you're a better man than I.

Try and be perfectly honest. Also be honest about how many mistakes are
there. By that I mean how many words come out on the screen wrongly compared
to the script you are reading.

> My speed claims are based on simple one syllable
> words. One-syllable words are a much fairer
> way of judging speed, in my humble opinion.
>
> The wpm metric means nothing unless one
> standardizes what one means by a "word".
>
> Better we junk wpm and use spm ;-)

Ahhh! I see now we are changing the goalposts. So now I have to count the
syllables and not individual words. I will try that only if you try my way
and just count the words. Just read from a letter or something different
from your regular scripts. Should be interesting.

> Now for 30 seconds, I have no trouble getting
> perfect 100% accuracy at 320 wpm.
> (one syllable words)

I believe I do see your nose getting longer here Mark and what use is
dictating one syllable words. It is unrealistic and tests nothing nor does
it prove anything. A typist would type a letter and would then count the
words he or she typed in a minute. They do not count the syllables of the
words, lol.

>> No Speech app at the moment is 100% word perfect
>
> You are confounding speech apps with speech app results.

So you agree, no speech app at the moment is 100% word-perfect!

> Are you saying that it is impossible to get 100%
> raw accuracy text results from a 7 minute
> dictation session?

I don't know, I haven't tried dictating for 7 minutes like that for a test.
I don't see the point! All I know is that it works well for me and is very
accurate. Yes, I still have a long way to go to learn different commands to
that of WSR, but from small acorns... as they say!

> If so, do not bet any money on it, because I consistently
> get such results.

You really do need to read my posts properly Mark. If you read one of my
earlier posts properly, you will have seen that I conceded that Dragon is
faster and more accurate than WSR at the moment on the dictation the side .
I have no argument there now.

I was talking about the 'command' side of things. Getting around the
computer faster. 'Open this', 'open that', 'Show numbers', select your
choice, no delay-instant with WSR. Much faster than Dragon at the moment.

Don't take my word for it, have a little look at that link I sent in my last
post. Majority use Dragon for dictating and WSR for navigation. That is
exactly what I do at the moment and have the best of both worlds. You said
this yourself if i remember rightly. All speech app's have their advantages
and disadvantages. I am just exploiting the best of each.

> You could also, if you get away from your present
> belief in the Magic Microphone, and concentrate on
> more important details such as advanced correction
> techniques, which Dragon has and WSR does not have.
>
> Keep in mind that I use a cheap headset that was included
> with Dragon, getting better resulting text accuracy than
> you do, even in the presence of loud blaring TV
> "infomercials", so loud that they hurt my ears.

I don't like headset or over the ear microphones. I am disabled remember so
I cannot take them off. Those days are over I am afraid. After a while
they become uncomfortable and start to hurt. That is why I use the
Sennheiser MD 431 ll desktop microphone which is super noise cancelling and
very accurate for my purpose.

I also said in one of my last posts that you obviously did not take in, that
one of the advantages of Dragon which I really love, is the fact that you
can have other things on in the background such as the television and you
can still dictate without it interfering. I am actually agreeing with you
here. Please read the posts properly and stop putting words into my mouth
that I actually did not say. You have a terrible habit of doing that you
naughty boy.

I also know about the 'Add Command' feature In DNS Pro where I can add my
own macro if I want. I also have the Knowbrainer software that gives me more
features on commands in DNS, thus making it even more accurate. I don't
believe you have this software, as you have never mentioned it before. Have
you?

> It would be amusing to see how you fare with that
> complex medical example, not very well I bet.

Not interested In anything medical . It doesn't rock my boat so is not
needed and doesn't present any test for me.

> You better rethink your methods, before you make
> any more of a fool of yourself by criticizing your peers.

I am just going by the facts as are plenty of other people who use WSR And
Dragon.

I still say quite confidently, I can get around the computer with commands
set in WSR quicker than I can in Dragon because it simply is faster. On that
score, I still say I would beat you around the computer better on the
command side of things if you use Dragon and I use WSR. For instance, I
have now switched to WSR for the sake of posting this post because I can cut
and paste along with all the other commands much quicker for this purpose.
I used Dragon to dictate in Microsoft Word first. To switch between the two
is easy peasy.

Eventually, my goal is to change the commands that are used in WSR to the
same as is used in Dragon so that they are in unison and to avoid any
confusion when I switch between the two.

> Thousands of hospitals world wide use Dragon.

So what? I am not interested in what other people use. I am only interested
in what works for me as a disabled person.

> I have not found even one hospital in this country that is
> using WSR, even though WSR has been around for years.

I don't care!

> This auxillary software can easily exceed $5,000
>
> Radiologists in particular _need_ this added software.

So what? I am not interested in what other people use. I am only interested
in what works for me as a disabled person.

Until Dragon can get rid of that awful delay between commands, catch up and
be as fast as WSR In that department, I will still use WSR for that purpose.

I doubt whether they will ever have a show numbers command as good as WSR.
Ask anybody who uses it. ;-)

That is not to say Dragon has its advantages in other departments as I have
already said three times now. :-)

Anyway, I see you have got what you wanted. You have started up quite a few
discussions about DNS once again ;-)

andy t

"Mark Conrad" <> wrote in message
news:271120092157131189%...

> What is all this fascination with speed?

Hey Andy, you are a post behind ;-)

> I packed up my mule, trekked out of these mountains to
> the nearest settlement (50 miles away) - did my banking,
> got a haircut, hit the general-store to pack my mule with
> provisions for the winter, fed my mule some pep pills
> for the long trek back up the mountain to my ranch.

>...frantically rushed into the ranch-house, expecting a reply
> to my recent post written on november 27, 2009.

> Nothing! - - - oh there was a post from you, but it was just
> a reply to an older post of mine, which I wrote on
> november 26, 2009.

Sorry that the anticipation disappointed you so, because it was obviously
something you were looking forward to.

Unfortunately Mark and believe it or not, I do have other interests in my
life and I do go out now and again. Yes, strange as it may seem, disabled
people do go out now and again. I am not one of those guys who sit over a
computer waiting for an answer to a post I sent. :-)

> Now I know that speed is not a priority for you, but consider
> that you would very likely have NO speech app to use,
> if speech apps were so slow that the rest of us would
> not buy them.

When you read my posts or anybody else's for that matter, do you actually
take any of it in?

Once again, this is for the last time Mark, so try and grasp it before you
die.

Dragon is faster than typing. WSR is faster than typing. No doubt, other
speech apps out there are faster than typing. So your first point is null
and void. So speed is already in the picture, it doesn't need to be a
priority for me because it is already there! As I have just said in
previous posts, I can dictate nearly twice as fast as a typist can type.
That's plenty for me and for most people I would expect.

Your second point is also null and void because you do not need to buy WSR!
It is already included it in an OS! Namely: Windows XP, Vista and 7.


> What is all this fascination with speed?

> In the Real World, speed is everything.

Well yes, to a certain extent but not to extreme and as I have already told
you, I can dictate nearly twice as fast as a typist can type. Have you
caught the bit between your teeth yet? Has the penny dropped?

> Without speed, people would use (ugh, cough) keyboards.

I cannot use the keyboard. I am disabled. I use speech recognition all the
time and I am faster than a typist.

> There would be no reason for people to create SR apps.

> Why should they? After all we have our keyboards.

The statement above is null and void! It has now been invented Mark. It is
like a nuclear bomb. You can't 'un-invent' something so, as to speak, once
it has been invented. Evolution has moved on from the keyboard to speech
recognition and speech recognition will keep on evolving just as everything
else does.

> The disabled community would be completely out of luck,
> because no one would bother to create SR apps for such
> a small market. If they did create one, they would have to
> charge $25,000 for it, to cover their costs of developing
> for such a small market.

WSR was made specifically with the disabled in mind. To add to that, they
are still a valuable part of the market that most speech app manufacturers'
enjoy. To add to that, your statement above is hypothetical and theoretical
and therefore does not belong. We shall never know!

> Speech Recognition is presently the fastest way of creating
> text from human thoughts, very important in any sort of
> time-critical human endeavor, where rapid widely-
> distributed text is required.

I agree, that is why I use it amongst obvious other things. Tell me
something I don't know! I also know you do not need to dictate like a rabid
dog to try and get as many words as possible on to the screen. Dictating as
you would normally talk to someone is fast enough and you are more likely
guaranteed not to have any mistakes by speaking at the pace. That is the
way you are meant to dictate. Ask any Dragon user or any WSR user. They
will all say the same. Dictate a steady pace, clearly and concisely for
best accuracy and fast results.

> If present SR apps failed to increase productivity because...

I had to snip the rest because of your constant, rambling and repetitive,
patronising statements. Besides, I have already answered them umpteen
times.

> Andy, Andy, Andy, I taught you everything I know, and
> now you know nothing ;-)

Um, correction. I learnt everything I know through reading, asking the
proper advice from experts and sheer grit determination! You have taught me
nothing yet Mark, the only thing you tried to do was to drop WSR and go
straight to Dragon. I am so glad I did not listen. Now I am ready for
Dragon and I will do exactly the same as I did with WSR, learn, learn,
learn!

> In my case, I need more speed and accuracy than you do.

> Bare minimum accuracy in my case is 99%, anything less
> causes me to revamp my technique.

I am getting exactly the same as you have finally admitted. 99% accuracy
and fast enough. Probably a long with all the other users as well. You are
not the unique, enigma that you like to see yourself as ;-)

As far as speed, the speed on everyday text, like your
80 word text below:

********************************
> I am using Dragon now and talking just at a normal pace
> and it is working perfectly well and is very accurate.
> This is the way it is meant to be used speaking clearly
> and concisely so that every word can be recognised.
> These were the exact words I was told by the guy
> tutoring me. I suppose you are now going to say he
> is wrong and you are right?
> I do not need to dictate at 211 wpm.
********************************

> That for me was a 25 second dictation.

> Everything went okay until I got to the very end,
> the darn MacSpeech stumbled, made a mistake.

> Should be:
> "211 wpm."
>...came out wrong, it printed:
>"211 words per minute." WRONG WRONG WRONG

When I dictated the letters "wpm" just like that, they came out first time
perfectly with DNS and WSR. Both programs have been trained that way. One
of my first triumphs :-)

>Where was I, oh yeah, talking about speed.

> Lemme see, 80 words in 25 seconds.

> That would be exactly 3.2 words in one second,
> or exactly 192 wpm.

Now now, stop keep moving the goalposts. You know that is not how the test
works or can be calculated by. Speaking of which, I see you did not reply
and pick up the gauntlet to do the test properly as I did it.

Dictate from a letter or something else that you have never seen before and
time yourself for 1 minute. Remember, this includes all punctuation but the
punctuation marks do not count as a word. Do not count syllables, count
words. A typist counts his or her words put down on the screen in a minute.
They also do not count punctuation or syllables.

That is a more realistic test. Then count the words afterwards. If you
have dictated 211 words in 1 minute and have NO mistakes on the page, then I
will bow to thee. I won't have to, because it ain't gonna happen!

Oh, when I say no mistakes, as I said before, the words you have dictated
from the letter have to be word perfect on the screen. Just try it, and I
will try your 'syllable challenge' and just count the syllables of words per
minute. In all honesty, that is not a proper test. It should be wpm
really.

> I refuse to go into Spelling Mode just to print "wpm".

I didn't need to :-)

I also noted there were a few questions I asked which you never answered.

More proof you do not read my posts properly. I get the feeling that you
are just interested in your own ego. Very rude of you Mark not to answer my
questions. :-(

andy t

"andy t" <andyt nospam.technet.invalid> wrote:

> "Mark Conrad" <aeiou mostly.invalid> wrote

>> Now I know that speed is not a priority for you, but consider
>> that you would very likely have NO speech app to use, if
>> speech apps were so slow that the rest of us would not buy
>> them.
>
> When you read my posts or anybody else's for that matter, do you
> actually take any of it in?

Maybe not, FumingTroll... Mark looks like an idiot
(for persistently trying to help you).
--




















> Once again, this is for the last time Mark, so try and grasp it before you
> die.
>
> Dragon is faster than typing. WSR is faster than typing. No doubt, other
> speech apps out there are faster than typing. So your first point is null
> and void. So speed is already in the picture, it doesn't need to be a
> priority for me because it is already there! As I have just said in
> previous posts, I can dictate nearly twice as fast as a typist can type.
> That's plenty for me and for most people I would expect.
>
> Your second point is also null and void because you do not need to buy WSR!
> It is already included it in an OS! Namely: Windows XP, Vista and 7.
>
>
>> What is all this fascination with speed?
>
>> In the Real World, speed is everything.
>
> Well yes, to a certain extent but not to extreme and as I have already told
> you, I can dictate nearly twice as fast as a typist can type. Have you
> caught the bit between your teeth yet? Has the penny dropped?
>
>> Without speed, people would use (ugh, cough) keyboards.
>
> I cannot use the keyboard. I am disabled. I use speech recognition all the
> time and I am faster than a typist.
>
>> There would be no reason for people to create SR apps.
>
>> Why should they? After all we have our keyboards.
>
> The statement above is null and void! It has now been invented Mark. It is
> like a nuclear bomb. You can't 'un-invent' something so, as to speak, once
> it has been invented. Evolution has moved on from the keyboard to speech
> recognition and speech recognition will keep on evolving just as everything
> else does.
>
>> The disabled community would be completely out of luck,
>> because no one would bother to create SR apps for such
>> a small market. If they did create one, they would have to
>> charge $25,000 for it, to cover their costs of developing
>> for such a small market.
>
> WSR was made specifically with the disabled in mind. To add to that, they
> are still a valuable part of the market that most speech app manufacturers'
> enjoy. To add to that, your statement above is hypothetical and theoretical
> and therefore does not belong. We shall never know!
>
>> Speech Recognition is presently the fastest way of creating
>> text from human thoughts, very important in any sort of
>> time-critical human endeavor, where rapid widely-
>> distributed text is required.
>
> I agree, that is why I use it amongst obvious other things. Tell me
> something I don't know! I also know you do not need to dictate like a rabid
> dog to try and get as many words as possible on to the screen. Dictating as
> you would normally talk to someone is fast enough and you are more likely
> guaranteed not to have any mistakes by speaking at the pace. That is the
> way you are meant to dictate. Ask any Dragon user or any WSR user. They
> will all say the same. Dictate a steady pace, clearly and concisely for
> best accuracy and fast results.
>
>> If present SR apps failed to increase productivity because...
>
> I had to snip the rest because of your constant, rambling and repetitive,
> patronising statements. Besides, I have already answered them umpteen
> times.
>
>> Andy, Andy, Andy, I taught you everything I know, and
>> now you know nothing ;-)
>
> Um, correction. I learnt everything I know through reading, asking the
> proper advice from experts and sheer grit determination! You have taught me
> nothing yet Mark, the only thing you tried to do was to drop WSR and go
> straight to Dragon. I am so glad I did not listen. Now I am ready for
> Dragon and I will do exactly the same as I did with WSR, learn, learn,
> learn!
>
>> In my case, I need more speed and accuracy than you do.
>
>> Bare minimum accuracy in my case is 99%, anything less
>> causes me to revamp my technique.
>
> I am getting exactly the same as you have finally admitted. 99% accuracy
> and fast enough. Probably a long with all the other users as well. You are
> not the unique, enigma that you like to see yourself as ;-)
>
> As far as speed, the speed on everyday text, like your
> 80 word text below:
>
> ********************************
>> I am using Dragon now and talking just at a normal pace
>> and it is working perfectly well and is very accurate.
>> This is the way it is meant to be used speaking clearly
>> and concisely so that every word can be recognised.
>> These were the exact words I was told by the guy
>> tutoring me. I suppose you are now going to say he
>> is wrong and you are right?
>> I do not need to dictate at 211 wpm.
> ********************************
>
>> That for me was a 25 second dictation.
>
>> Everything went okay until I got to the very end,
>> the darn MacSpeech stumbled, made a mistake.
>
>> Should be:
>> "211 wpm."
>>...came out wrong, it printed:
>>"211 words per minute." WRONG WRONG WRONG
>
> When I dictated the letters "wpm" just like that, they came out first time
> perfectly with DNS and WSR. Both programs have been trained that way. One
> of my first triumphs :-)
>
>>Where was I, oh yeah, talking about speed.
>
>> Lemme see, 80 words in 25 seconds.
>
>> That would be exactly 3.2 words in one second,
>> or exactly 192 wpm.
>
> Now now, stop keep moving the goalposts. You know that is not how the test
> works or can be calculated by. Speaking of which, I see you did not reply
> and pick up the gauntlet to do the test properly as I did it.
>
> Dictate from a letter or something else that you have never seen before and
> time yourself for 1 minute. Remember, this includes all punctuation but the
> punctuation marks do not count as a word. Do not count syllables, count
> words. A typist counts his or her words put down on the screen in a minute.
> They also do not count punctuation or syllables.
>
> That is a more realistic test. Then count the words afterwards. If you
> have dictated 211 words in 1 minute and have NO mistakes on the page, then I
> will bow to thee. I won't have to, because it ain't gonna happen!
>
> Oh, when I say no mistakes, as I said before, the words you have dictated
> from the letter have to be word perfect on the screen. Just try it, and I
> will try your 'syllable challenge' and just count the syllables of words per
> minute. In all honesty, that is not a proper test. It should be wpm
> really.
>
>> I refuse to go into Spelling Mode just to print "wpm".
>
> I didn't need to :-)
>
> I also noted there were a few questions I asked which you never answered.
>
> More proof you do not read my posts properly. I get the feeling that you
> are just interested in your own ego. Very rude of you Mark not to answer my
> questions. :-(
>
> andy t
>
>
>
>

> From: "andy t" <andyt nospam.technet.invalid>
> References: <251120090648451513%aeiou mostly.invalid> <#7alrPrbKHA.2160 TK2MSFTNGP02.phx.gbl> <261120091725365181%aeiou mostly.invalid> <#qD5g05bKHA.1652 TK2MSFTNGP05.phx.gbl> <271120092157131189%aeiou mostly.invalid>
> In-Reply-To: <271120092157131189%aeiou mostly.invalid>
> Subject: Re: Confession for my nasty abuse of the Dragon speech engine
> Date: Sat, 28 Nov 2009 13:18:18 -0000
> Lines: 7
> MIME-Version: 1.0
> Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=original
> Content-Transfer-Encoding: 7bit
> X-Priority: 3
> X-MSMail-Priority: Normal
> Importance: Normal
> X-Newsreader: Microsoft Windows Live Mail 14.0.8050.1202
> X-MimeOLE: Produced By Microsoft MimeOLE V14.0.8050.1202
> Message-ID: <Oq7nI1CcKHA.4688 TK2MSFTNGP06.phx.gbl>
> Newsgroups: microsoft.public.windows.vista.general
> NNTP-Posting-Host: 79-74-146-38.dynamic.dsl.as9105.com 79.74.146.38
> Path: news.astraweb.com!border1.newsrouter.astraweb.com! news-xfer.nntp.sonic.net!news.alt.net!msrtrans!TK2MSFTF EEDS02.phx.gbl!TK2MSFTNGP01.phx.gbl!TK2MSFTNGP06.p hx.gbl
>
>
>

Mark Conrad <> wrote:

....

> Tsk, tsk, tsk - you certainly have a very hard tme grasping the
> most simple concepts, as evidenced by this remark of yours,
> concerning the simple fact that you would not have WSR to play
> with if the rest of the able-bodied world refused to buy speech
> recognition apps.

Too true.

I have always wondered why speech recognition was not included
with Windows (and argued for it), considering the fact that input
is a fundamental part of computing and speech input is no
different than keyboard input for people who cannot type.
Surprisingly, counter argument came from some pseudo-techy Linux
Lunatics, saying "You want everything put into Windows!"

The same applies to speech activated scripting, that is using your
voice to tell a computer what to do. In Windows 3.1 long ago,
Microsoft touted Macro Recorder as a disability aid. But before
Windows 98 came along, Microsoft pushed Macro Recorder out the
back door. Fortunately there was third-party macroing software.
Someday (hopefully) we will even see speech activated scripting
that includes Continuous Command Recognition as part of Windows,
unless Microsoft thinks we should pause between every command
given to a stupid computer (as with WSR). Continuous command
recognition is not rocket science, I am doing it proficiently here
and now. Unfortunately, Windows moves like a quadriplegic (not
bashing Windows, Mac and Linux might be worse).

Someday in the future, after hearing enough pause-and-command
instruction, the computer replies... "Why are you talking like
that?"

Mark Conrad <> wrote:

> In article <008d3860$0$13097$>, John
> Doe <> wrote:
>
>> ...that is using your voice to tell a computer what to do.
>
> Yeah, I often wondered why a user is presently required to pause
> both before and after saying a command.
>
> It should be obvious to the stupid computer that we are asking
> it to do something.
>
> Maybe computers can't do the obvious?

Dragonfly does Continuous Command Recognition, works great here. I
have a large list of mostly Windows shortcuts as one or two
syllable utterances (plan to keep a reasonably current list in the
group comp.lang.beta).

Vocola included CCR up until DNS 8. I thought it was rocket
science, but found out that it just requires a small amount of
programming. Dragonfly works with both DNS and WSR, but
unfortunately, you almost have to be a real programmer in order to
get Dragonfly going with DNS, and WSR is a pain to work with.

Windows Speech Recognition could easily include CCR, but
apparently not enough users are interested in systemwide control
of their PC via voice, and Microsoft programs for the masses.

Continuous Command Recognition opens up the ability to string
commands together instead of using predefined phrases. That makes
a huge difference for syste mwide voice activated scripting in
Windows. Seems obvious to me that will be the way instructing a
stupid computer/robot will be, but high-tech moves very slowly
especially when the leadership is mainly interested in holding
monopolies...

Mark Conrad <> wrote:

> John Doe <> wrote:

>> Dragonfly does Continuous Command Recognition, works great
>> here.

> Could a 3rd-party developer create a simplified CCR add-on for
> regular users who do not want to program?

I am sure that a Python programmer could, Yes. The author can
answer that question clearly. Dragonfly is supposed to be for
Python programmers. The author of Dragonfly would like to simplify
its installation for end users, but he has only so much free time.

Dragonfly is written in Python and it is open source, here.

http://dragonfly.googlecode.com/svn/...ion/index.html

> Could you explain, or show an example, of how CCR can benefit an
> ordinary user?

It is much like the difference between the way speech recognition
started and the advent of Continuous Speech Recognition. That is
why I use the expression "Continuous Command Recognition". BTW,
some prefer the phrase "command sequences".

The main benefit (to me) is much greater proficiency, you can
combine commands just like ordinary words in a human language.
With pause-and-command, I had to make phrases for everything
because pausing between individual command words would be a waste
of time. With CCR, you make a list of commands as if you were
making a list of words for a human language, and then you can
combine them into phrases on-the-fly just like people do when
speaking. A short list of words is very concise compared to a long
list of phrases. And then adding one word can add much to the
language since it can be combined with other words. Without CCR,
using phrases, you would have to add that single additional word
to every phrase you might use.

So... Write some simple single-syllable-activated scripts to
replace Windows shortcut keys and mouse clicks (and some actions
that are not Windows shortcuts, but maybe should be), like the
following...

"word" = select word under pointer (double-click)
"get" = copy selected text
"switch" = switch windows

Now you can point and script (the pointer is used a lot, but most
clicking is automated).

The phrase "word get switch" without pausing grabs the word under
the pointer and takes it to the other window. I use that
frequently when doing stuff like window shopping or research, for
moving words between windows.

Add these...

"line" = select line under pointer (starts with a single click)
"put" = paste text from clipboard
"go" = press Enter

Now, the concise phrase "line put go" replaces the line under the
pointer with the contents of the clipboard and presses the enter
key. That is also very useful for stuff like window shopping or
research.

Those two examples use the words in the sequence that they are
listed, but that is not necessary, you can mix and match all six
words anyway you like with CCR.

Commands could be multiple syllables, but I have little trouble
using single syllables. Whenever a single syllable gets in the way
of doing anything else in Windows, it gets changed to use some
other syllable (or, if absolutely necessary, two syllables are
used).

Just like with ordinary scripting, many words in my vocabulary
include an optional number suffix for the number of times the
key/action will be repeated. Using Dragonfly, you can do command
sequences with or without using those suffixes for any of the
individual commands in the phrase.

"Mark Conrad" <> wrote in message
news:281120090905391392%...




> Tsk, tsk, tsk - you certainly have a very hard tme
> grasping the most simple concepts, as evidenced by
> this remark of yours, concerning the simple fact
> that you would not have WSR to play with if the
> rest of the able-bodied world refused to buy
> speech recognition apps.

But I have got WSR and DNS to play with so your question is hypothetical and
therefore null and void!

Next question?

> In 1970's....

<SNIP irrelevant drivel >

Stop living in the past accept that the Wright brothers are dead.

> Gawd, you are dense!

Says an 80 year old who still cannot spell properly :-)

Sometimes I can be, I grant you. But I have DNS Pro + KnowBrainer 2008 +
WSR + Win7 and I am doing quite well thanks. So who's the dense one really?

> WSR would NEVER have been developed.

But it has! Get over it and look to the future.

All this flak trying to blind everyone with science and not once mentioning
the test or I set you. You remember that don't you?

You know, the one you keep changing the rules with. Surely you have not
forgotten already? Or is it just selected dementia? ;-)

> You are too dense to realize every handicapped
> customer, there are 100 able-bodied customers.

And your point being? God, you do go off at a tangent and into an utterly
pointless ramble.

If you haven't any family left, I suggest you find some friends and have a
chat with them instead of yourself all the time.

>Anyhow, I finally got my buggy MacSpeech to instantly
>output "wpm" whenever I spoke the letters wpm.
>(I did not need to resort to a text macro)

What took you? I have no problem with that in either app. And it is you
that is crappy not the software.

Anyway, I thought you were dealing with syllables per minute and not words
per minute. You keep changing your mind so much that I personally think you
do not know what you are doing.

Why don't we count the letters per minute? Then you might get somewhere near
the made up figures that come into your head all of a sudden.

> Now I need to get back to work, to make certain that I can
> dictate those 80 words at least 3 times at around 200 wpm
> without any text mistakes, to justify using SR technology.

Yes okay you do that. Can you say hello to Alice in Wonderland while you're
doing it?

> After all, without all that speed, I might as well revert
> to ordinary typing.

keep taking the speed and happy dreams! ;-)

andy t

Mark Conrad <> wrote:

> John Doe <> wrote:

>> So... Write some simple single-syllable-activated scripts to
>> replace Windows shortcut keys and mouse clicks (and some actions
>> that are not Windows shortcuts, but maybe should be), like the
>> following...
>>
>> "word" = select word under pointer (double-click)
>> "get" = copy selected text
>> "switch" = switch windows
>>
>> Now you can point and script (the pointer is used a lot, but most
>> clicking is automated).
>>
>> The phrase "word get switch" without pausing grabs the word under
>> the pointer and takes it to the other window.
>
> Aha, now I get the idea, thanks.
>
> Looks like I will have to hit the B & N bookstore and buy a
> book on the Python language.
>
> Should not be too painful. Back in my productive days when I
> still had a brain, I was a fair to middling Lisp programmer.

The author definitely can use some help implementing the thing for
end users.

You do not have to be a Python programmer to get Dragonfly working,
but the current installation procedure is a daunting task.

I just junked together an explanation of how to install and use
Dragonfly (at least in Windows XP SP3), including hopefully all of
the necessary files, uploaded "Dragonfly.zip" to (comp.lang.beta).

Good luck.















>
> Sheesh, that is all I need, another project :-\
>
> Here, I have one foot in the grave, and my other foot
> on a slippery banana peel, so my remaining time on
> planet Earth is limited.
>
> Mark-