Voice of Vision
Volume 2 Number 1 -- January 1994

This issue of VV is loaded with helpful information for PC users.  Have
you ever used your exception dictionary to alert you to misspelled
words?  Elmer Slentz has an idea that you might like to try.  If you've
ever used TEXT2SET and SET2TEXT (formerly VE_CMPL and VE_SET),
or if you don't know what they're for, Mike Lawler's article will tell you
about new features and give you some hints.  Jeff Pledger continues
his three-part series on memory management.  Jay Leventhal
announces the American Foundation for the Blind's on-line information
bank.  Doug Geoffray tells us what Dragon Dictate and IBM VoiceType
are and helps us decide whether or not voice input is for us.

_______________
EXCEPTION DICTIONARY AS TOOL FOR TEACHING SPELLING?  WHY
NOT?
By Elmer D. Slentz

Have you ever thought that your very own exception dictionary could
be used as a teaching tool for children and adults?  Within the next few
pages, we will discover a little bit of how this can be done.

I currently use the Sounding Board speech synthesizer, which, of
course, uses the exception dictionary with which many of you are no
doubt familiar, as it changes the pronunciation of various words to your
liking.  That same dictionary can be used to teach spelling as well.  Let
me give you an example:

Load your favorite text editor such as Noteworthy, or your favorite
word processor, such as WordPerfect, and then create a simple ASCII
text file.  You might give it a file name like SPELLING.DIC, and save it
on disk in your Vocal-Eyes directory.  Now, on the first line, for
example, type the word CAT and press enter.  On the next line, type
the word again followed by a colon and then skip a space or two. 
Then type the letters each with a space between them, as C A T, and
again press enter.  Now you have the first entry of this new spelling
dictionary.  Notice the first line has just the word itself; then the
second line contains the word followed by its spelling:  "CAT:  c a t". 
(Quotation marks are simply used to enclose appropriate examples). 
Now save it as you would any other exception dictionary.  Now use
your command line to load your newly created spelling dictionary,
using the same procedure as you always have.

          SB /ESPELLING.DIC

If your dictionary loaded successfully, try typing the word CAT, and
see what happens.  Your computer should have pronounced the word
and then spelled it immediately afterwards.  There you are.  It is just as
simple as that.

I use this technique on my job as medical transcriptionist to assist me
in correcting words which I find difficult to spell.  The word
"Darvocet," for example, is one that I had quite a bit of difficulty
learning to spell, so I finally placed an entry in my dictionary.  I placed
that word on one line and then put "THAT IS CORRECT" on the second
line.  That second line is the line that normally gives you the corrected
pronunciation of a given word, but this time it clues me in that I spelled
the word correctly.  However, you need to be sure that you have the
correct spelling before using that technique, because your spelling
dictionary is only as accurate as the one who enters the word into it. 
Then I created other entries in which I deliberately misspelled the word
on one line and then on the second I typed "CORRECT SPELLING:  D A
R V O C E T."  Then I loaded the newly revised dictionary, and every
time I would type that misspelled form of the word, it would give me
the correct form so that I could go back and retype it correctly.  I
decided to think of the various ways that I might misspell the word,
and enter each one into the dictionary, each as its own entry, with the
second line correcting the spelling.

Now, this technique eats up a lot of memory very shortly because we
are using misspelled words as the base for giving ourselves the proper
spelling, and, as you know, each word is an entry in itself and takes
two lines.  Therefore, if you have possibly three misspelling entries for
the word "Darvocet," and anywhere from two to maybe five
misspellings for several other words with which you have difficulty,
you can see how quickly your dictionary would run out of space, even
with its enlarged capacity to hold up to something like 25K of RAM
memory.  I am hoping someday soon that the exception dictionaries
will not have a limit except for the limit set by the RAM memory
available, since medical terminology has so many words that need
pronunciation and spelling assistance.  I believe the technology is
available to turn your exception dictionary into a teaching device for
spelling purposes, or  even use it on the job to correct words with
which you have difficulty in spelling.

Below is a sample of a spelling dictionary, exactly the way it would be
written to be used as such.  the words being dealt with are "relevant",
"with", "gas", "aspirin", and "harbor":

relevant
that is correct.
relavent
correct spelling:  r e l e v a n t
relavant
correct spelling:  r e l e v a n t
with
that is correct.
witn
correct spelling:  w i t h
witg
correct spelling:  w i t h
gas
that is correct
gass
correct spelling:  g a s
aspirin
that is correct
asperine
correct spelling:  a s p i r i n
asperyn
correct spelling:  a s p i r i n
harbor
that is correct
harber
correct spelling:  h a r b o r
harbur
correct spelling:  h a r b o r

This is merely a sample, but I used it in my own computer and was
able
to load this particular dictionary.

It can even assist with typos from time to time if they happen to use
any of the misspellings that you entered into your exception dictionary. 
Why not try it some time and see how it goes?  Then if you care to,
please get back with me and let me know how it turns out.  I'm a
frequent user of the GW Micro BBS, so give me  a call and leave me a
message there.  The phone number is (219) 484-0210.   Please let me
know if you find this dictionary usage of benefit.

_______________
AFB'S CAREERS AND TECHNOLOGY INFORMATION BANK
by Jay Leventhal

Would you like to speak to a blind or visually impaired person who is
employed in your field or in a field you are interested in going into?
would you like to share information, with other users of assistive
technology, including over 100 users of GW Micro products?  The
Careers and Technology Information Bank (CTIB) offers first-hand
information about job experience and assistive technology from people
who are blind or visually impaired,

The CTIB is a network of 1500 blind and visually impaired people who
use assistive technology at home, at work, or at school.  It provides a
unique opportunity to exchange information and get advice.

You can get in touch with someone who is willing to discuss how they
perform their job tasks, or exchange information with another user of a
particular product.  CTIB members have also agreed to serve as
mentors for high school and college students, and will offer advice
about career choices.  A complete list of jobs held by the people in the
CTIB is available in print or in braille upon request.

To participate in the CTIB, or to tap into its resources, contact;

careers & Technology information Bank
American Foundation for the Blind
15 West 16th Street
New York, NY 10011
Call the CTIB at (212) 620-2080.

_______________
Vocal-Eyes users already know how to use Steve Holmes' two handy
utilities SET2TEXT and TEXT2SET (formerly known as VE_SET and
VE_CMPL).  You've used SET2TEXT to create readable lists of your
SET file settings and TEXT2SET to change the rate in all your SET files
or in a group of them.  Steve, the author of these two utilities, and
Mike, GW Micro product support, have gotten together and made some
substantial improvements in these utilities.  You can now do more than
change your menu settings or redefine your hot keys.

How to make the most of SET2TEXT and TEXT2SET
by Mike Lawler

Have you ever wanted to modify cursor positions, window attributes,
or hyperactive window commands within several set files?  Well, now
you can do all of that and more with version 1.6B of SET2TEXT and
TEXT2SET.

TEXT2SET can now compile every Vocal-Eyes feature.  It also has a
new option called "cursor_key" that allows the user to make global
changes to the cursor keys.  For example, lets say that you wanted to
undefined the backspace key in every set file, but you didn't know
where in the cursoring key list the backspace might appear.  You can
now use the syntax

     cursor_key = keyname command [command]

So for our example you could say

     cursor_key backspace = undefined

and TEXT2SET would find the backspace key in each set file and
undefine it.  If the backspace was not defined in a set file TEXT2SET
would simply ignore the command.

Another improvement to TEXT2SET is duplicate key checking. 
TEXT2SET will not allow you to assign a hot key twice or a cursoring
key with more than one function (except, of course, LINK).

SET2TEXT also has been improved.  It now lists the set file name and
the last modification date and time of the set file as a comment in the
text file.  It also lists the version number of SET2TEXT that created the
set file.  SET2TEXT now supports wildcards.  If you wanted to
decompile all of the WordPerfect set files you could use the following
syntax

     SET2TEXT WP*.set

and SET2TEXT would decompile every WordPerfect set file to a text
file with the same name, but with an extension of ".TXT."

SET2TEXT and TEXT2SET now only require one command-line
parameter.  The following commands would be equivalent:

     SET2TEXT VE VE

     SET2TEXT VE

will both decompile "VE.SET" to "VE.TXT."

Finally, TEXT2SET has been modified so that it doesn't update your
SET file unless there are no errors in your text file.  If you have errors,
you'll receive a message like

Error C:\SPEECH\FILE.TXT 1: Repeat filter must be OFF, SAY
REPEATS, or IGNORE
C:\SPEECH\FILE.TXT was not compiled to C:\SPEECH\ABC.SET due to
1 error
0 out of 1 SET files were compiled.

Just correct the problem and try again.

We think that these enhancements will make your use of SET2TEXT
and TEXT2SET both easier and more enjoyable.  You can find version
1.6B of these utilities on our BBS in the Vocal-Eyes file area under the
name "VEUTILS.ZIP."  Our phone number is (219) 484-0210.

_______________
What About Voice Input?
by Doug Geoffray

Some of you may or may not be aware that Vocal-Eyes actually comes in
two flavors.  We have the standard version and we also have the
Dragon/VoiceType version.  Here I'll discuss DragonDictate-30K, IBM
VoiceType and the Dragon/VoiceType version of Vocal-Eyes.

Approximately two years ago I was contacted by National Easter Seals to
see if Vocal-Eyes could be use with DragonDictate.  The DragonDictate
and IBM VoiceType are large-vocabulary voice recognition products. 
These allow you to talk to your machine instead of typing at the
keyboard.  You can dictate a letter to your machine instead of typing it. 
Almost everything that you can do from the keyboard can be done by
talking to your machine.  The DragonDictate allows for 30,000 words
and the IBM VoiceType allows for 7,000 words.  To put this into
perspective, IBM did a research study which showed the average college
graduate's vocabulary is 7,000 words.  Thirty thousand  words is as
many words as Shakespeare used to create all his works.  You can also
add proper names, jargon, or  any words or phrases suited to your
specific needs.  This is an important advantage if you need a customized
vocabulary for a specific job or industry.  Both systems are so flexible
that you can replace all 30,000 or 7,000 words with your own.

How does it work?  Let me pull the following description from the
DragonDictate literature.

     DragonDictate's high performance is based on its recognition
     of phonemes and subphonemic acoustic elements, the building
     blocks of speech.  Essentially, DragonDictate identifies these
     speech blocks, groups them into words, and then compares
     the words against word models stored in its program. 
     Powerful statistical and mathematical algorithms analyze the
     phonetic features of a word to determine its correct match.

     DragonDictate recognizes separately spoken words or phrases. 
     Each time you say a word, a box appears on screen with a list
     of words that DragonDictate thinks you said, ranked in order
     of probability.  If the word you said appears at the top of the
     list, you simply continue dictating, giving DragonDictate the
     go-ahead to enter the word.  If your desired word appears
     farther down the list, you select it by voice or keyboard.

     If you say a word, and none of the possibilities on the choice
     list match what you said, you just begin spelling by voice
     using the program's spell mode, or typing it in.  Once
     DragonDictate knows how the word starts, the program tries
     to recognize it again, usually getting it right after you say or
     type one or two letters.  Or, you can edit a word on the
     choice list that is closest to the one you said.

     The power behind DragonDictate is its ability to adapt to your
     voice, vocabulary, and work environment.  When you first use
     DragonDictate, it creates a personal voice file for you that
     contains words and pronunciations for both the active
     vocabulary and the backup word list.  As you dictate and
     make recognition corrections, DragonDictate personalizes your
     voice file based on the particular way you speak.

There is much more to DragonDictate but this should give you a good
idea of what it can be used for.  The IBM VoiceType acts exactly the
same as the DragonDictate with a smaller vocabulary of 7,000 words. 
IBM has licensed the technology from Dragon Systems.

Now, back to the Easter Seals question of, "Will Vocal-Eyes work with
DragonDictate?"  They were attempting to accommodate a blind
individual who was physically unable to type.  He could not use the
keyboard and could not see the screen.  We tried the standard
Vocal-Eyes and had limited success.  It required some keystrokes in
certain areas and simply would not speak the relevant information in
others.  The only way to get the two to work together was to modify one
of the two programs.  Although it would have been a dream come true
for Dragon Systems to modify their software, reality was knocking me on
the head.  We decided to create a DragonDictate/VoiceType aware
version of Vocal-Eyes.  This is a special version and must be specifically
asked for when placing an order.  The special version can be made to
work as a standard
version but not the other way around.

The special version of Vocal-Eyes constantly looks for Dragon/VoiceType
information to appear on the screen and speaks it in a meaningful
fashion.  For example, as you dictate each word, the word list pops up
on the screen.  Visually, you can look at the first entry and, if it is
correct, you simply dictate the next word.  With Vocal-Eyes, it will
automatically read the first or
highlighted word in the list.  If you like, you can instruct Vocal- Eyes to
spell the word after a predetermined amount of time.  This comes in
handy for homophones.  Therefore, as you say each word, Vocal-Eyes
will read the word DragonDictate thought you said.  If you pause, the
word will be spelled.  If the word is correct, you simply dictate the next
word and so on.  If the word is not the word you said, you can do one of
two things.  You can manually search the word list to see if the word is
there but not as the first option.  Or, you can start spelling out the word
you really said.  The second approach, although it sounds more tedious, I
recommend.  Because DragonDictate has a backup dictionary, you
usually only have to say the first couple of letters before the word you
said is placed as the first option in the list.  This is great for me since I
am a horrible speller.  Each time you spell out a new letter Vocal-Eyes
will read the new word if it has changed.  If you want it to, Vocal-Eyes
will also spell the new word giving you the current cursor position within
the word.

With the special version of Vocal-Eyes, the .SET files have been modified. 
For example, the keyboard voice has been turned off.  You don't want to
hear each letter as Dragon sends the word to the underlying application. 
Also, since you can instruct Dragon to delete characters, we have
undefined the backspace key in the cursoring menu so you don't hear
each letter as it is being deleted.  We have tried to make the
Dragon/VoiceType interface as automatic as possible.

Is the DragonDictate and Vocal-Eyes combination for everyone?  In my
opinion, not at all.  With DragonDictate alone, I have reached speeds of
about forty to fifty words per minute.  If I use DragonDictate with
Vocal-Eyes and do not use the monitor, the best I have ever done is thirty
words per minute.  However, the average speed in the field is about ten
to fifteen words per minute.  If you previously could not use a computer,
ten words per minute seems infinitely fast.  If however, you can type at
speeds exceeding ten words per minute, dropping down to ten words per
minute may not be acceptable.  DragonDictate is a great choice if you
have mobility problems, have a learning disability, or if you have carpal
tunnel syndrome.  If you also have vision difficulties, adding Vocal-Eyes
with DragonDictate allows for a very unique way of communicating with
your computer system.  If you would like to frustrate your sighted
friends, simply turn off the monitor, push the keyboard aside, put on a
headset, lean back in your chair and dictate away.

_______________
An Introduction to Quarterdeck's QEMM
by Jeff Pledger

I stated in my first article that I would be doing a product evaluation of
Quarterdeck's Expanded Memory Manager (QEMM) version 7.0.  Since
the writing of the first article, Quarterdeck has come out with three
maintenance releases for this product.  The latest version is 7.03.  When
and if you are thinking of buying this package, I recommend that you
purchase a full package with Quarterdeck QEMM, Manifest and
Desqview.  This article will focus on the general installation of QEMM and
a very brief summary of optimizing your configuration.  Later articles will
go into further detail on optimizing your system and utilizing the utilities
which Quarterdeck offers.

As I said in my first article, memory is distributed into four areas: 
conventional/low (first 640kb), expanded/EMS, upper memory, and
extended/XMS memory.  What QEMM allows you to do is to optimize the
conventional memory area where you run your DOS applications by
optimizing where your Terminate and Stay Resident (TSR) and specific
device drivers are located.  These TSRs and device drivers will be loaded
into the upper memory area.  At a minimum, your computer system
needs the following to run QEMM:

     a) IBM/compatible  model PS/2 or A/T with a 80386 or higher micro
     processor      
     b) A minimum of 2Mb of RAM; however, if you have 8 or 12Mb you
     will be optimally set for running QEMM.
     c) Running either Microsoft or IBM DOS version 3.3 or higher.
     d) Approximately 1.5 to 2Mb of free hard disk space.

I state these are the "bare bones" requirements as I became familiar with
them and have worked with Quarterdeck's products for the past four
years.  Now that we understand the requirements, lets move on to the
installation and primary set-up of the CONFIG.SYS and AUTOEXEC.BAT
files.

To install the Quarterdeck products, place the installation disk #1 in your
floppy disk drive and type either "A:INSTALL" or "B:INSTALL."  During
the installation, turn on the Vocal-Eyes read box feature.  This will read
the screens as they change.  If you have a "Read Box" hot key set, this
can prove helpful when rereading the screen after it has spoken to you. 
The installation process is extremely simple.  Just follow the prompts for
the default settings for installing QEMM, manifest and Desqview.  We will
go into further detail on truly optimizing your system in later articles, but
for now, lets just get the system loaded and working with the default
settings.
As I said, you should load QEMM, manifest and Desqview during the
installation process.  Make sure that you answer the questions correctly
during the install to do this.  You can not just copy files from the floppy
disk onto your hard disk with the "copy" command.  You must use
Quarterdeck's installation program.  This is the only way to ensure a
complete and proper installation.

Once you have completed the installation correctly, you will be prompted
to run "OPTIMIZE"or to "exit to DOS".  You should run optimize to gain
the full advantage of your computer.  Just follow the prompts as they
appear on the screen.  You can do either an express mode optimize, one
key stroke to perform all of the stages, or you can examine what occurs
during each stage of the optimization process.  Remember to follow the
prompts on the screen.  Optimize works in several stages: the analysis,
detection and final phases.  I will briefly explain the meaning of each.

The analysis phase is where OPTIMIZE copies the contents of your
AUTOEXEC.BAT and CONFIG.SYS to holding files and analyzes what
programs are loaded during the computer boot-/start-up procedure.  You
are prompted to strike any key when this is completed.  Again, this
program is very friendly to use and it will give you a prompt when it
requires action from you.

The next phase is the detection phase and it starts with a re-boot of your
computer.  This phase will detect what programs will fit into the Upper
memory available on your system.  If you are loading, lets say, the
following programs, you should have no problem with memory allocation:

  a) Vocal-Eyes with a speech synthesizer driver
  b) a disk caching program like SMARTDRV.EXE or PCTOOLS PC-
  CACHE
  c) an anti-virus program
  d) a driver card for a scanner

If, however, you are loading many TSR programs such as LAN drivers or
other peripheral device drivers, you may get a message that all of your
programs could not be loaded high.  You will have the option of running
something known as "STEALTH-ROMS."  This will make available more
upper memory by using a "paging scheme" to buy the additional memory. 
You should follow the prompts on the screen to enter the "STEALTH"
process.  It would be advisable to have someone sighted with you for
this process.  QEMM will do some system checking to see which
"STEALTH" parameter will work best on your system.  This is where you
will need the assistance of your sighted friend.  You should follow the
prompts on the screen as the computer re-boots.  You will have to put a
formatted floppy disk in your disk drive, usually drive A.  Since the
computer has just re-booted and Vocal-Eyes has not yet been initialized
you will not have speech at this point.  Your sighted companion will have
to read the prompts on the screen for you.  At this point, the "STEALTH"
process will detect which parameter will work best with your system. 
Once this process is done, you will be returned back to the detection
phase for completion of its part in the process.

The final phase saves the configuration made during the optimize process
permanently to your hard disk.  This is not to say that you can only
optimize your system once, oh contraire, you can optimize as many times
as you wish.  Once the optimize has completed successfully, you will be
returned to DOS.  You can examine what has been loaded into Upper
memory by issuing the LOADHI command at the DOS prompt.  This
program shows a mapping of how TSRs and device drivers were loaded
high into the Upper memory area.  When the system has come back to a
DOS prompt, you should examine your CONFIG.SYS and
AUTOEXEC.BAT files in the root directory, c:\>, and see what
changes/enhancements were made.  One line I will discuss in particular is
the QEMM.SYS line of your CONFIG.SYS.  This line should be either the
first or second line in your file.  It should read like the following:

  DEVICE=C:\QEMM\QEMM386.SYS RAM [EXCLUDE=XXXX-YYYY]
  [ST:{M/F}] R:#

The above device line says the following device driver (QEMM386.SYS )
will be loaded into the upper memory region marked through the R:#
parameter.  Three other QEMM command line parameters are used on the
same line.  They are as follows:

  A) RAM
  B) EXCLUDE (X)
  C) ST:[M/F]

To use a QEMM386.SYS command line parameter you type the
parameter name on the same line as DEVICE=QEMM386.SYS in your
CONFIG.SYS file. It is important that you do not put any spaces within
the parameter syntax, e.g. X=DC00-DEFF.  Each QEMM command line
parameter has an abbreviation which you may use instead of the
parameter name.  The abbreviation for each QEMM386.SYS command
line parameter, if applicable,  is shown with in parentheses after each
parameter name.

Here is a brief summary of what some of the QEMM command line
parameters will do with QEMM386.SYS.

RAM or RAM=XXXX-YYYY puts RAM everywhere or in the range
specified. The RAM parameter must be set if you want to run TSRs,
drivers or DOS resources in high memory.

EXCLUDE=XXXX-YYYY, considers the range specified unmappable. 
This means that QEMM will not be able to map TSRs or device drivers to
the range specified.  You would use this for loading something like the
Calera Truescan recognition card for the Arkenstone Reader version 2.x. 
you would use this parameter in the following manner:  X=DC00-DEFF. 
Note there are no spaces at all within the syntax of the parameter.

ST:(M/F), Stealth ROM may seem mysterious, but it really is not.  What
the STEALTH ROM offers to you is the ability to load system ROMS and
other ROMS  into high memory through either a mapping scheme (ST:M)
or through the page frame (ST:F).  The ST:F does not offer as much of a
memory gain as using the ST:M parameter.  You really do not have much
control over this parameter at this point in our discussion.  If your
computer will accept the ST:M parameter use it instead of the ST:F .  A
problem arises in that not all computers are compatible with the ST:M
parameter, so you are forced to use the ST:F parameter.  This may be
further complicated by your system telling you that you can not use
either of these parameters, thus taking away the ability to use STEALTH
ROMS.  Another drawback to using STEALTH is that your system slows
down due to the extra processing it has to perform.  This difference is
more noticeable on a 386 than on a 486 processor. If you don't have to
use this parameter, then my recommendation is not to.

I strongly suggest that, if you're considering purchasing QEMM, you
purchase Recording for the Blind's (RFB), QEMM, Manifest and Desqview
manuals in the IBM Bookmanager format.  This file viewing program is
designed by IBM to work with an electronic media format for books. 
Vocal-Eyes has set files written which work very well with this program. 
It would be most advantageous for anyone who wishes to know about
using Quarterdeck's QEMM, Manifest and Desqview.  I will also be
referring to the manuals in later articles for QEMM command line
parameters and the manifest utilities.  The book # is EP030 and the cost
is $15.00.  This book will only give you the manuals for Quarterdeck's
QEMM and Manifest version 6.  All version 6 QEMM command line
parameters and Manifest utilities are supported in version 7.x.  There is
an additional charge for owning the IBM Bookmanager program.  For
information on these services call RFB at (800) 221-4792.

Well, I guess I will close out this issue of memory management for this
session.  I would give an overall rating of using QEMM of a "10" on a
performance enhancement to your system.  It is extremely friendly with
speech with little configuration needed.  Hopefully, many of you will
purchase the QEMM386 manual I noted earlier in this article.  It will help
for future references which I will be making.

_______________
Hints for New Users
by Crista Shaikh

This regular column is intended to help users who are new to the
products discussed or to computers in general or who simply feel
they need more help than the manuals they have read have provided.

Question:  My new application isn't in graphics mode, but it still isn't
talking.  I'm told a menu is on the screen, but when I press the arrow
keys I don't hear anything meaningful.

Answer:  There are two things you can do right off the bat to help make
any new program talk.  They probably won't make it speak perfectly, but
you'll be eighty-five percent of the way there.

First, turn Auto Box Read on.  Many programs display relevant
information in boxes.  The steps would go like this:

1)        Get into your new program.
2)        Pop up the control panel by pressing CTRL-BACKSLASH.
3)        Go to the screen menu, which is number one and press ENTER.
4)        Go down to Auto Box Read, which is number sixteen if you're using
          Vocal-Eyes 2.2 or number thirteen if you're using Vocal-Eyes 2.11.
5)        Turn it on.

The other thing you can do is turn light bar tracking to Auto.  As long as
you're in the control panel, go to the hot keys menu and define a bar
toggle hot key.  The steps will go like this:

1)        Go to the main menu, either by pressing ESCAPE if you're still in the
          screen menu or by popping up the control panel with CTRL-
          BACKSLASH.
2)        Go to the Hot keys menu (number three) and press ENTER.
3)        Go to option fifty-eight.
4)        Press the key you'd like to use as your bar toggle hot key.

Now go back to your application (press ESCAPE twice) and press your
new hot key until it says "AUTO."  Try moving around the menu with the
UP, DOWN, LEFT, or RIGHT arrow key.  After one or two presses of the
correct arrow key, you should hear the menu item as you go.

I have made these two changes and have saved them in a file called
INSTALL.SET.  This way, whenever I install a new program, it talks
automatically, without my doing anything special.

Question:  What is a TSR?

Answer:  TSR is a term you've probably heard used in computer manuals
or thrown around by your hacker friends.  It stands for Terminate and
Stay Resident.  This is applied to a computer program (software) that is
run by the user or by another program and then stops (terminates) but
stays in memory.  If it is in memory it is said to be resident.  PC users are
already familiar with this concept because a screen reader is a TSR.  You
run the program, or your AUTOEXEC (what's that?) runs it for you, and
when it finishes doing its initial job, it stays active.


General Hints

Have you recently upgraded to Quatro Pro 5.0?  You can  persuade it to
cooperate better with your keyboard if you use the /k parameter provided
by Borland.  To get into the program, you would do something like this:

  Q /K

This looks similar to a Quicken 5.0 hint in a recent issue of VV, but as far
as I know, the resemblance is coincidental.

Do you use Procomm Plus or Telix for telecommunications?  Maybe you
have a problem with getting text from a BBS or other utility to be read by
Vocal-Eyes without having to hit a hot key or use some hyperactive
window scheme.  Both of these programs have commands to solve the
problem.

If you Use Procomm Plus 2.01, you'll find it on the set-up menu (ALT-S). 
On this menu you'll find an option called DISPLAY/SOUND OPTIONS.   In
older versions of PC Plus it's phrased differently, but you'll recognize it. 
Press enter on this option and you'll get a list of options which can be
selected by pressing a letter between A and K (or some other range if
you're using an older version).  Select the option "SNOW REMOVAL,"
which is choice E if you're using version 2.01.  Use the space bar to
toggle between YES and NO.  Press ENTER when you have it on YES. 
Use the ESCAPE key to go back to the Set Up menu and select "SAVE
SETUP OPTIONS" to make your change permanent.

In addition to the above change, I created a simple set file called
PCPLUS.SET.  I turned autobox read on, light bar track to auto, set
window zero to lines one through twenty-four and window one to line
twenty-five only.  That's it.

If you're using Telix, your changes will be similar to the ones described
above.  Go into the ALT-O, "Configure Telix," menu and select "Screen
and Color."  You'll be presented with another menu from which you'll
select "Screen write mode."  Telix will give you three choices and a brief
explanation of them.  Choose "2."  You can also choose "3" if you like,
but if you do everything will talk.  The main place you'll find that to be a
problem is in the menus.  Whichever you choose, be sure to return to the
"Configure Telix" menu and save your settings.


Telix also has a -b parameter which will give the same effect as the third
choice on the menu described above.  The syntax would be

  TELIX -B

Create a TELIX.SET file identical  to PCPLUS.SET, and the job is done.

Do you use MegaDots to create and print Braille documents?  If you use
it with Vocal-Eyes or one of several  other screen readers, be sure to get
version 1.3 if you haven't already.  In older versions, your system would
lock up if you attempted to use the VE cursor in MegaDots.  MegaDots
1.3 solved that problem.

Also on the subject of MegaDots, did you know that program has some
special settings for speech users?  Go into the program with the /V and
/U command line parameters.  The line will read:

  MEGA /V /U

The /V parameter tells MegaDots that you're using voice and the /U
parameter tells it Unframed text editor preference.

You can also save these setting so you don't have to type them every
time you enter the program.  Press F10  and then "P" "S" for
"Preference," "Save."  Enter a new file name if you like and then press
ENTER or just press ENTER to accept the name already there.

Do you have a monochrome system?  Many applications rely heavily on
color to indicate important items in menus or to display error messages. 
If you're using a SET file made by a friend, you'll probably need to tell
Vocal-Eyes to pay attention to blacks and whites instead of the cyans,
magentas, and greens that your friend  told it to watch for.  We've done
SET files for monochrome configurations of several popular applications,
in addition to color versions.  If you'd like monochrome SET files for
WordPerfect 5.1, WordPerfect 6.0, or Quicken 6.0, these are all on the
bulletin board.  Also, if you've upgraded recently, they are on your
upgrade disk in a directory called SETS.  If you've copied all the new SET
files from your upgrade disk and are in your speech directory, simply type

  MWP
  MQUICKEN

To cause the files to self-extract.  MWP contains the SET files for
WordPerfect 5.1, WordPerfect 6.0, and WordPerfect 6.0A.

_______________
Announcements

Vocal-Eyes Upgrades

If you're a registered Vocal-Eyes user, you were sent a card a few
months ago letting you know that VE 2.2 was out.  If you bought VE
during 1993, that upgrade is free.  If you didn't get a card it may be
because you've moved and we don't have your new address, or it may
be that you forgot to register when you bought the program.  In either
case, please call or write and let us know whether you need 3.5- or
5.25-inch disks.  Please be sure to include your current mailing address
and your VE serial number (this is located on the help menu of the control
panel).  If you bought Vocal-Eyes before 1993 and you have version 2.0
or greater, the cost of the upgrade is only $20.00.

After February 1994, the free upgrade offer will end and those upgrades
will also cost $20.00

Video tape announcement

GW Micro is pleased to announce that we now have a video tape which
demonstrates many of our products.  It is available to teachers,
counselors, disabled student service organizations or anyone else that
needs to see how our products are used in "real-world" applications.  Not
only is it interesting visually but it is well described for those who cannot
see the video and it is closed captioned for those who cannot hear the
audio.  Our video illustrates how speech output can be utilized by people
that are visually impaired or (with the help of visual tracking in Vocal-
Eyes) have a learning disability.  It also demonstrates the DragonDictate
products with and without speech output.  If you or someone you know
has a need for this video, please contact us and we will gladly send you a
copy.

DragonDictate starter edition

DragonDictate Starter Edition is the first speech-to-text dictation software
for PCs (speech input) to be priced below $1000.00.  Until now, the
lowest priced dictation program cost several thousand dollars.  The
DragonDictate Starter Edition is $595.00 for the software and $375.00
for the hardware bringing the total price to $970.00.  The program
comes with a 5,000 word active vocabulary containing the words used
most frequently in everyday communications.  In comparison, the
DragonDictate-30K has a 30,000 word active vocabulary and is priced at
$2295.00.  The DragonDictate Starter Edition makes large vocabulary
dictation much more accessible to all PC users.  If you can't, or don't like
to type, give us a call about this exciting new product.

Four-track tapes

The four-tracks are coming!  The four-tracks are coming!  In response to
numerous requests, we have decided to switch to four-track recordings
of our tape catalog and instruction manuals.  If a product has a small
manual we will probably not change it.  However, products with several
tapes will be soon changing to four-track.  This allows us to use fewer
tapes and in the case of our catalog, put more information on one tape.

New Prices

Dragon Systems, developers of IBM VoiceType and Dragon Dictate, have
announced a new price for their high-end voice input product.  Dragon
Dictate II, the thirty-thousand-word voice input system, was previously
priced at $4995.00.  As of January first, 1994, the price for this product
will be $2295.00.  We think this makes this product accessible to many
more people who might have thought they would find voice input
beneficial, but who couldn't afford it.

SET files Available

If you don't like to make SET files, you should check GW Micro's bulletin
board.  Many new groups of files for Vocal-Eyes users have been added. 
Recently, we've added MS Works, WordPerfect Works, Professional
Write, and several more DOS 6.X utility configurations.  Download them
all or only the ones you think you'll use.  The  phone number is
(219) 484-0210.  If you've logged on before, you'll notice we have a
new files area dedicated to SET files.

Thanks

Special thanks to Larry Fontan of West Virginia for his help in formatting
and designing the Braille version of this newsletter.  If you have any
comments on the Braille you've received, or if you'd like to change your
subscription from another format, please call us.  Subscription
information is listed below.
_______________
Subscription Information

Voice of Vision is published four times a year in large print, on audio
cassette, on 5.25- or 3.5-inch PC diskette, and in Braille.  You can
receive a subscription for $10.00 for each format.  VV is also available
on GW Micro's bulletin board at no charge.  Payment can be made by
check, MasterCard, Visa, American Express, or Discover.  Send checks
and format specifications to the address below.

We seek reader input.  If we use an article you've submitted,
you'll receive a years subscription at no charge.

Please send articles on IBM compatible diskette or in print to:

GW Micro, Inc.
310 Racquet Dr.
Fort Wayne, IN  46825

or upload them to our bulletin board at (219) 484-0210.  Articles may be
edited for brevity and clarity.

If you have any questions or comments, please contact us at
(219) 483-3625.
