EASI Archives

Equal Access to Software & Information: (distribution list)

EASI@LISTSERV.ICORS.ORG

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Ramon Castillo <[log in to unmask]>
Reply To:
* EASI: Equal Access to Software & Information
Date:
Mon, 9 Jul 2001 09:03:23 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (81 lines)
Lots of questions.  I'll start by saying that speech recognition can indeed
be useful for people with LD.  As for the Voice Profile becoming worthless,
just don't save the speech files and it won't get worse, or make a backup of
a good supervised speech files and replace the user file when it gets bad.

It seems that if the job just requires him to write about what has been
fixed on cars, the vocabulary and phrases used will be limited and similar,
greatly increasing the likelihood of decent accuracy.

We have used speech recognition with text-to-speech programs for people with
severe dyslexia.  I also had a client who would verify the dictated text
with a Franklin Language Master, when in doubt.  This is especially helpful
with homonyms which can be missed when using a text-to-speech program to
check dictation.

For some clients, DragonDictate or VoiceDirect has been more effective,
given that it is easier to correct one word than a phrase.  Also, the list
after each word is dictated is very helpful and allows the user to pick a
word if they have decent word recognition ability.

For clients with non-verbal learning disabilities, speech recognition has
been especially helpful.


--
Ramon Castillo, OTR, ATP
Occupational Therapist, Assistive Technology Practitioner
ATEC/Goodwill Industries of Orange County
200 North Fairview
Santa Ana, CA 92703
714-547-6301  ext.389
Fax  714-541-2387
http://www.ocgoodwill.org/atec <http://www.ocgoodwill.org/atec>




-----Original Message-----
From: Martin, Terry [mailto:[log in to unmask]]
Sent: Monday, July 09, 2001 8:42 AM
To: [log in to unmask]
Subject: Voice recognition


Greetings all:

I am looking for your opinions on recommending voice recognition for
learning disabled students.  First I am looking for what you think of this
tool overall for LD individuals.  Does it work?  What kind of success have
you had with it?  What kinds of LD individuals are using it?  What are your
success stories and what are your horror stories.  I am not comfortable
recommending it especially for those who cannot read or write.  It seems
like the wrong tool.  I do understand the thinking that it is better to get
something out of him than nothing.  It may work at first, however, if they
cannot read or spell well enough to correct the recognition errors,
eventually, their voice profile is going to be worthless.  Then, they will
have to start over again.  Is this productive?  Will a state agency pay for
retraining again on the same product?  What does everyone think?

I do have a specific case where a 23-year old cannot read or write.  He
claims that he can recognize some words.  He is very smart with automobiles.
The employer likes him very much.  The evaluator is recommending Naturally
Speaking.  The evaluator does not like voice recognition because they do not
think it works well.  I will probably end up doing the training.  If this
does not work, the person will probably lose the job that he has right now.
I have not talked with the person yet myself.  This is all that I know for
the moment.  The person's job requires him to write reports on what was
fixed on the car that his team worked on.  What are your thoughts?

Thank you for the assistance.
_____________________________________________________
Terry Martin -- President
VOILA Technology, Inc.
54 Castle Road
Rochester, NY 14623
Phone:  (716) 321-1451
FAX:  (716) 334-3971
Email:  [log in to unmask] <mailto:[log in to unmask]>
URL:  http://www.mainaccess.com <http://www.mainaccess.com>
____________________________________________________

ATOM RSS1 RSS2