Windows Vista Tips

Windows Vista Tips > Newsgroups > Windows Vista General Discussion > Mark Conrad is still lying about his speech dictation speed!

Reply
Thread Tools Display Modes

Mark Conrad is still lying about his speech dictation speed!

 
 
andy t
Guest
Posts: n/a

 
      02-12-2010


"Mark Conrad" why is he incapable of telling the truth?






Interesting link that you posted about speech recognition. And you are
right in saying I was right in saying speech recognition has only been worth
trying within the last four years or less.



>You claim that I posted that I could dictate at 600 wpm,
>which I never posted.




Yes you did. You posted a link to a video on You Tube that showed the
fastest talking woman in the world. She can speak 600 words per minute,
which is true, she can.


Unfortunately, she was speaking so fast that to any human, it sounded like
gobbledygook and NOBODY in the world could understand it. It took a very
special machine to convert what she was saying to text (NOT DRAGON).


You implied the software they were using for recording her speech was Dragon
or you implied that Dragon could decipher her words and put them on screen
and be 100% accurate. This would have been impossible! I cannot remember
which one of those two statements it was but people in here can easily type
in your name, look back over your old posts and see I was telling the truth
and that you were lying and making things up as usual.


It is there in black and white for everybody on the web to see. I'm not
going to let you wriggle out of that one Mr wriggly squiggly snake.



>You claim I backpedal and revise my dictation speed




Yep! You are renowned for it!



>because whenever you
>make such out-of-context exaggerations, you are very
>careful not to quote the context of my original statement.




Bullshit! The above statement is another example of how this cretin tries
to wriggle out of a situation when he realises how ridiculous his statements
sound after-the-fact.


All his text has been put down in black and white complete including the
entire context around it for everybody to see on the web. Anyone can easily
check back to your earlier posts and see I was telling the truth about you.



>The fastest speed I ever managed to dictate at with 100%
>accuracy was and is 336 wpm, using the standard test for
>typists, namely counting ALL keystrokes, then dividing by 5,
>then adjusting for 60 seconds. (One minute)




BULLSHIT and WRONG!!! (WHAT **** DOES THAT MEAN ANYWAY???). Explain
yourself Mr wriggly squiggly snake??


The standard test for typing you are on about is carried out nothing like
you have described above. What you have written is complete tosh as per
usual.


The standard word count is where they count blocks of five keyboard strokes
and then each block of five is counted as word. Not what you said above.


Even doing it that way, you would be nowhere near the imaginary word count
you have dreamt up above or indeed any of your other imaginary words per
minute count that you have mentioned before.


Read this link in Wiki to prove exactly what I am saying and for the
standard test. In this test, every block of five keystrokes is counted.
Even calculating with this test, you would get nowhere near your ridiculous
claims. You would probably get 150 words per minute if you were lucky with
no errors.


http://en.wikipedia.org/wiki/Words_per_minute


Firstly, nobody mentioned using any standard test. All I said was count the
words per minute. This is the normal test for typists.


Not syllables, not characters, not divide by this or divide by that and not
multiplied by the left hypotenuse of a diameter on a chimpanzees arse!


Just count the words per minute not including spaces or punctuation
(although you still have to put these in, the same as a typist would have to
but they are not counted toward your total) and then afterwards take off the
number of word mistakes you made from the your dictated words totalled.


THEN TELL US HOW MANY WORDS YOU DICTATED IN ONE MINUTE AND HOW MANY MISTAKES
YOU MADE? Can't make any plainer than that!


Look at this link. They show you how to calculate wpm


Http://www.ehow.com/how_5133296_calc...er-minute.html



This is exactly what they say and how it should be prepared and undertaken.
It's quite easy to follow really. Surely, even you can take this in with a
step-by-step guide.


Step 1: Prepare the text you will type. Take the text from a book or
magazine. Do not use a text that you are familiar with or you will not have
accurate results.


Step 2: Set the egg timer for one minute. As soon as you set it, you are
ready to type.


Step 3: Type the text until the egg timer sounds. At this time, you will
count the words or use "Tools" in Word to find the count.


Step 4: Proofread the text and count up your mistakes. Take the total
amount of words typed and subtract the mistakes for your total word count
per minute.


If you do your dictating in Word, there is even a word count in the bottom
right-hand corner so it does it all for you. That is the word count and
that is the number you should be telling us less the word mistakes you made!


My daughter-in-law took her typing test and she told me this is exactly how
they test you by doing it this way.


Okay, I am going to make it even easier for you to take the test and anybody
else out there can try this, either by typing or by using speech
recognition.


Below is a link that will take you directly to a typing test that will test
your speed in either typing or speech recognition. Once you get to the
website, you just click 'Start Typing Test' and you just type or dictate
their test document for one minute including all punctuation and spaces.


You don't even have to time yourself because they do it all for you and at
the end of the minute you don't even have to count the words or the errors
because they do that all for you too.


THEN come back and tell me how many words per minute you can dictate and how
many errors there were. This test could not make it any simpler for you to
test your speed!


Here is the link: http://www.now1technologies.com/typetest




>The standard test for typists is for TYPISTS, it is far from
>being an ideal test for speech dictation, but despite this obvious
>fact, it is widely used when speaking of speech recognition speeds.
>The most accurate way of measuring SR speed is to merely use
>prepared text, in other words POST the text that is being used
>ahead of time, along with the TIME it took to dictate that text,
>along with the NUMBER of mistakes, if any.




Bullshit! You do not dictate naturally like that by dictating pre-arranged
text from one source over and over again! Above is a prime example of you
moving the goalposts again and changing the rules to suit your needs Mr
wriggly squiggly snake! Read the link above that takes you to Wikipedia and
THAT will tell you what the standard test for typing speeds really is. EVEN
USING THAT you will get NOWHERE NEAR your fictional speeds!



Just take the typing test from the link I show above THEN tell me how many
words per minute you dictate and don't forget, this includes all punctuation
and spaces and don't forget to tell us how many errors you make as well, as
these will be taken off the eventual word count.




>For example, I posted a 600 word test that took me 240 seconds
>to dictate, which was 100% correct in 3 out of every 4 tries...





FORGET your examples. That is you making up your own rules again. In the
typing test link I show above, they give you a test document there to
dictate from, without any of your medical words. There is no need to make
any macros, as all the words are standard. Just dictate that.


The most I managed was 154 words per minute without any errors. I could
speak more by speaking faster and get over 160 words per minute but it
brought up errors because by speaking that fast, the software could not
understand and therefore decipher certain words I was saying.




>That boils down to 150 wpm as measured by the standard typing
>test, with no mistakes in average of 3 out of 4 tries. I posted ALL
>the macros I used, there were 4 very small macros, addresses of
>medical supply companies.





Once again, that is your interpretation of the typing speed test and your
own made up crap. THAT does not boil down 150 words per minute. THAT boils
down to bullshit! Take the typing test as shown above THEN tell us how many
words per minute you can manage without any errors!




>The word usually missed was always "transesophageal". I am still
>working on a way to improve accuracy when dictating that
>particular word...





Now, you are telling us of a word that is usually missed and you are still
working on it. This sort of blows your 100% claim out of the water, doesn't
it. Backpedalling again!




>so I can jack up my pathetic 150 wpm speed
>to a higher level. the word "transesophageal" spoken at any speed
>above 150 wpm is asking for trouble, right now.





Now we are saying 150 words per minute. That is a long way from 600 words
per minute you originally claimed Dragon could succeed with and being 100%
accurate.


I have already told you in numerous posts before that I have managed just
over 150 words per minute without any errors WITH BOTH WSR and Dragon. So
this is not unique and can be achieved. So what's new?


Remember, this guy started saying that you could dictate 600 words per
minute using Dragon. Then realising what an idiot he made himself look like
by saying this, he changed it to 300 words per minute. Then he changed it to
syllables per minute because he must have looked at it again, then tried
doing it, realised it could not be done and quickly backtracked by changing
it to syllables per minute. Realising by saying this, made him look like a
pratt. He swiftly changed it to 260 words per minute. I called his bluff
once again because he was talking bullshit as usual. He must have tried his
darned hardest to get 260 words on screen and not to expect any mistakes but
obviously failed miserably (because it cannot be done). So he comes back and
then changes it to 200 words per minute and expect there not to be any
mistakes, which is STILL Bullshit! As I pointed out to him.


So he has obviously had another rethink and bit his tongue as you can see
above in black and white because now he is now admitting 150 words per
minute. Don't get me wrong, that is still fast and I have been saying this
is more realistic all throughout these shenanigans of his. But it's a long
way from what he first professed @ 600 words per minute and changing his
tune FIVE times along the way!


What a dork! And he's still trying to wriggle out of it and justify it
today.


No one believes what you say any more Mark . It's just words from a
bull-shitter.


Even Dragon themselves admit that the average speed that someone speaks is
120 words per minute. Your imaginary claims of some 300 words per minute, is
180 MORE words per minute on top of that. Utter bullshit!


http://www.dragonvoicerecognition.co...ofessional.cfm



Look under the heading: "Features for Dragon NaturallySpeaking professional
solutions."



.... And then just underneath the subheading: "Up to 99% Accurate and Three
Times Faster than Typing.



.... Read how many words per minute Dragon reckons most people can speak!




>Half Truths




You don't know the meaning of the word truth! You only know what is inside
your head, which is fairytale bullshit and tall stories!




>Although you agree that the WSR feature is COMPLETELY broken
>in the base WSR application, and according to your own words





By 'the feature', I assume you mean the 'reviewer' option in speech
properties. In this case, yes I have agreed it does not work and explained
why in one of my previous posts.




>everyone knows that it is broken, you still strongly imply that "WSR
>Toolkit" fixethat, when actually >it does not fix it.




I have never implied that the toolkit has a reviewer in it. So if there is
no reviewer in the toolkit, how can it be broke?


What I have said and posted in detail in one of my previous posts, is that
the add from file feature when used in conjunction with two other features
within the toolkit, namely the add to dictionary and train from text
features, will improve accuracy in dictation. Which it does.




>The only thing "Add from File" does, in WSR Toolkit, is to add
>NEW names and words, that are not yet in the users speech
>vocabulary.





No it does not. I posted in detail in one of my previous posts exactly what
the add from file does along with what the add to dictionary does in detail
along with what the train from text feature does in detail. I also tell
that THEY ALL should be used in conjunction with each other to get the best
effect for greater accuracy. It is NOT the case that the add from file
feature just adds words to the vocabulary.


Read my previous posts.

This is what the add from file feature actually does.


'Add from file feature'


This feature will allow you to improve the accuracy of your "Language
Model." It's not the same as adding words to your personal dictionary but
works by taking the text that you pass it, parses it into words, and then
records the relative frequencies of occurrence of each word with respect to
the words around it. If you use documents that represent your style of
speaking, this feature will improve dictation accuracy.


Currently only MS-Word 97-2003 (.doc) files, MS-Word 2007 (.docx) files and
Text (.txt) files are recognized.


Choose 'Browse' to locate your Word .doc file, then choose 'Begin Add from
File' to begin.


This is not a 'reviewer' or 'document scanning feature' as you call it.


That is not the same as adding new names and words, which are not yet in the
user's speech vocabulary.


Read my previous posts.




>is does not improve accuracy at all, even if you scanned hundreds
>of documents .




When used in conjunction with the other features in the toolkit, yes it does
improve accuracy and I didn't need to use hundreds of documents, I only used
roughly 80. AND it does not improve accuracy the same way as a reviewer
does.


Read my previous posts



>You already admitted that you can only get 98% to 99% average
>accuracy,




Only get 98 to 99%??? What do you want blood? That isn't bad considering it
was significantly lower before I had the toolkit and for a piece of software
that only cost $15.99.


>especially on complex text, where you likely only
>get 98% or less.




I have explained how to add as you call it 'complex text', words or phrases
that are not in the vocabulary by simply, adding them to the dictionary and
recording a pronunciation of that word or phrase.


Read my previous posts!


Hepaticocholangiocholecystenterostomies.


You see. Still there from when I trained it and you can't get more complex
than that! I will admit, I had to remember how I trained and pronounced it
exactly but seeing as I will never ever use it, I will not lose any sleep
over it ;-).


There you go again putting words in my mouth. I did not say and I have never
said "98% or less." I have said all along, I am getting 98 to 99% accuracy
using dictation in WSR. That isn't bad considering it is free. Includes
the show numbers command which eventually you will have to buy on top of the
Dragon software (and it's not as good as WSR's version anyway). Does not
need the KnowBrainer software to enhance the amount of commands it has
(which is essential in Dragon and will cost you another $200). Comes within
a whole operating system anyway. By buying a piece of software costing
$15.99 will enhance accuracy with the features it has in it. Will enable you
to make text and command macros, which cannot be done in the 'Preferred'
version of Dragon and considering it has only been going just over two years
compared to Dragon's 18 years.


The reason I mention 98 to 99% accuracy in WSR is because I have already
admitted and I agree that Dragon is slightly more accurate using dictation.
I am constantly getting 99% accuracy all the time in there. No matter what
you say, NO SPEECH APP IS 100% ACCURATE using dictation. Ask anybody or any
expert and look at the link Re: Dragon above. Even they admit 99% accuracy
is their peak.


I have told you many times that I use both because WSR is easily faster at
getting around the computer than Dragon on the command side. Dictation is
just one part of a speech app. The command side is equally important. I see
you conveniently forget to mention that part when you come back with your
bullshit full on dictation ;-). Because YOU KNOW, I am right and brush it to
one side.



>I can easily get 100% accuracy...




No you cannot. Nobody can because no speech app is 100% accurate using
dictation. Stop making things up again and telling Porky pies Mr wriggly
squiggly snake!



>On complex medical text...




Don't need complex medical text ;-).



>and you
>could also, if you correctly used the: Tools -> Accuracy Center
>of your Dragon Pro, looking for the following section:




I already am and do ;-). Tell me something I don't know ;-).



>"Add words from your documents to the vocabulary"




Yawn! I already do and have. Try and keep up young Mark ;-).


< SNIP > the rest. You really are stuck with old news aren't you.




>and expect 100% accuracy.



Bullshit!



>If you try the same routine with "WSR Toolkit" it
>just plain will not work, no 100% accuracy...




98 to 99% accuracy I get using the WSR toolkit. 99% accuracy all the time I
get with Dragon Professional and NO APP IS 100% ACCURATE with dictation
contrary to your bullshit above Mr wriggly squiggly snake.




>this $1,600 app performs badly on general dictation
>of everyday common language, a fault well known to all
>who try to use the medical version for everyday SR use.




In that case, what a waste of money it is.


You would have been better off purchasing Dragon Professional and adding the
medical words to the vocabulary because Professional has no problem with any
words apart from certain homonyms and certain small ones that sound the same
as I have mentioned before.



>enlighten yourself about the importance of a
>working document scanning feature.




Durr! I already know about the reviewer in Dragon. I have used it several
times.


It is an important feature but not imperative in the toolkit as explained
umpteen times why it is not.



Hepaticocholangiocholecystenterostomies. Look at that ;-).


Came out first time again proving you do not need a reviewer for the
software to remember that word. ;-) ;-)


Don't forget to try your dictation speed in the typing speed test link as
shown above. And no cheating and coming back telling Porky pies.


I want to know how many words per minute and how many errors you made. See
if you can tell the truth for once ;-).


andy t


 
Reply With Quote
 
 
 
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Why does Mark Conrad keep putting his foot in his mouth? andy t Windows Vista General Discussion 5 02-09-2010 04:44 PM
Re: Why does Mark Conrad live in a world of his own??? andy t Windows Vista General Discussion 0 02-03-2010 11:37 AM
Re: Why is Mark Conrad so thick? News Forum Check Windows Vista General Discussion 0 01-31-2010 03:10 PM
Mark Conrad needs a new brain transplant andy t Windows Vista General Discussion 0 01-30-2010 03:17 AM
Re: Speech Recognition, Pros and Cons andy t Windows Vista General Discussion 0 12-23-2009 09:49 PM



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59