Tip:
Highlight text to annotate it
X
This video tutorial will teach you how to train your computer to better understand you
using the speech recognition feature on a Windows operating system.
This technique will come in handy
when you need to do your audio captioning for Camtasia.
A little bit of preparation on your part
will go a long way to making your video editing experience better.
You will need to login the user as yourself before you begin the speech recognition program.
I have set up a user as Camtasia Recording on my home computer.
The speech recognition feature in Windows 7 can be found by going to the Start Button -->
Control Panel -->
Ease of Access -->
Speech recognition.
I'm going to choose the topic that says "Train your computer to better understand you".
This dialog box will get you some instructions
on preparing for the Speech Voice Recognition training.
Please read the following text out loud:
I am now speaking to my computer
The computer is learning the sound of my voice as I speak.
This will help the computer better recognize what I say.
Speech recognition can recognize many speakers very accurately
but training the computer will provide an even better experience
there will be fewer mistakes when dictating
and when I get voice commands such as
correct that
go to the end of the sentence
and start wordpad
the system will be better able to understand what I say
some advice for speech recognition users
by now you have probably already used speech recognition a little
there are many ways to become more proficient with speech recognition
one way is to make sure the computer here's you
you should practice articulating well
configured speaking like a newscaster on television or on the radio
that style of speaking usually works well for speech recognition
many dictation mistakes are caused by these are not speaking clearly
ensure also at the microphone is always positioned correctly
that boom on a headset microphone should be 1 inch away from your mouth
otherwise the sound of your voice will not be recorded well
think of when a friend called you on the telephone
but did not speak directly into the fall
you probably had a hard time
understanding what was said
if you think you may have a poor quality microphone
consider using a better microphone
another way to improve your dictation
experience is by fixing mistake efficiently
the collection dialogue is designed to make correcting text easier and faster
if the computer recognizes further at father
do not select father
followed by sank further
instead just say correct father
you'll probably see the word you want in the list at Pierce
if you don't see the word analyst just speak a word again
if you can't still can't see the word after you say it again
you can fake spell it
the spelling dialogue is very patient and powerful
unit 88 is for apple or a avid Adam
you could also say three to add the letter three
or in fake delete delete the current letter
and a fake insert to insert a letter
just say OK when you get the word you want
the computer will learn from the collection
the computer will be unlikely to make the same dictation mistake again
another way to improve your dictation experience is to learn advanced commands
if there are two instances of the work apple in the document
and use a file like apple
you will feed that each instance of the world apple had the box over it
to select a word you want fate the number of the word followed by OK
if a menu appears and you want to get rid of it
simply fake cancel to dismiss the menu
to school down the page in a document just a scroll down
if you don't know the name of the problem you want to switch to
just say switch application
you can then choose the one you want from the list on the screen.
This concludes that it's for speech recognition training session
A little bit of preparation on your part
This concludes that it's for speech recognition training session
to read more about tips
look in the help documentation for speech.
This concludes the speech recognition training
if you want to continue to train your computer you would select the More Training (button).
I will go ahead and select the More Training
and I will be directed to more text to read.
speech is the main way people bought and learn from each other
the way we speak affects the way we can indicate with each other
on the telephone in movies and on radio television and the Internet
supported by the mouse and keyboard
speech is slowly becoming one of the primary ways we interact with computers
and technology that allows us to control computers via speech is still very new
especially when you consider that
the ultimate goal of speech recognition technology
is to make talking to a computer as natural as
talking to another human being
this requires the computer to both be able to understand human speech
and to simulate human speech
understanding human speech is called speech recognition
simulating human speech is called text to speech speech conversion
when people first started making speech recognition software
they decided to take a large numbers of incoming phone calls
this was a good place to start
because recognizing the callers speech meant
this was a good place to start
because recognizing the callers speech meant
recognizing only a few different responses
each suggested by the software
text to speech conversion
also worked well in telephone systems
the system usually have only had to simulate short precise
braces
short and precise phrases are easier to pronounce
for text to speech conversion systems
personal computers are now
beginning to do a good job for speech recognition
and text to speech conversion
kick personal computers allow users to complete a wide variety of task
but for all their future virtues computers understand speech
differently and human being
computers understand words by analyzing sound waves
and they understand sentences by judging how likely it is
that certain words will appear next to each other
the more different things a computer have to listen for
the harder it is to accurately recognize what a person might say
pick up
on the telephone the caller is expected to respond to a prompt
with concise information
in this case the file for usually knows exactly what to listen for
such as an account number
of zip code or departure city
using a computer people can do many things at any given time
we can even do many things simultaneously
we can chat with a friend
and read a news article at the same time
since the software doesn't know what we're doing a lot will do next
it has to assume that we might say just about anything
at any given time
this challenge becomes even more complicated when people dictate to the computer
the computer has to be ready to recognize anything we might say
even if it has nothing to do with what we just said
or were not dictating but still talking
personal computers handle this problem
by dividing what ain't what a user might say
in command and dictation
the computer limits the number of command we can speak at any given time
the computer lessons only for small set of commands
that can apply to what we're doing
when we're working on something that allows us to dictate text
the computer still tries to interpret what we say as a command
if the computer doesn't recognize what we say at the command
it will interpret the speech as dictation
the computer then enters this dictation
into the current program as text
this concludes the speech recognition background information training session
to learn more about speech recognition
look in the help documentation for speech
I could continue to take
selecting a more training button
but I'm going to stop at this juncture
and I'm going to select next
upgrade estate deal sent
I'm going to close out of my control panel