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:
Alan Cantor <[log in to unmask]>
Reply To:
* EASI: Equal Access to Software & Information
Date:
Thu, 5 Jun 2003 00:24:43 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (49 lines)
Speech recognition software is complicated. It is not as straightforward as
many people think. Operating a PC by voice is not at all like using a
Dictaphone or tape recorder. Successful use of programs like NaturallySpeaking
hinges on having the discipline to learn -- and apply -- proper dictation
techniques, a good understanding of how the program works, and an intuitive
grasp of language. That's a tall order for a ten-year-old. I don't know many
children of that age who would be able to use speech recognition
independently.

I have introduced NaturallySpeaking to two, gifted, computer-savvy children,
ten- and twelve-years-old. Both got through the enrollment, and both were able
to use the program in limited ways -- but only with constant adult supervision
and support. Left to their own devices, the children treated the software like
a game. It was more fun to repeat the misrecognized words that appear on the
screen, say their name again and again, and to squeal, yell, or speak another
language. Both got very frustrated dealing with misrecognized words,
unintended utterances, and spurious microphone sounds.

I would not categorically rule out the possibility that a ten-year-old child
can (1) learn NaturallySpeaking and (2) actually use it to write, but I would
not get my hopes up too high. I question the educational value of teaching a
child to use speech recognition software if he or she is does not learn to
produce fairly accurate text. The ability to produce correct text is related
to the degree to which the software is prepared, and the user's ability to
recognize misrecognitions AND correct them. (Anyone can produce word-soup
using speech recognition software, but I do not consider the resulting text to
be "writing!")

Ultimately, having the right kind of adult support should increase the chances
of a successful implementation of speech recognition software when working
with a child.

Alan

Alan Cantor
Cantor + Associates Inc.
Workplace Accommodation Consultants
[log in to unmask]
www.interlog.com/~acantor

> The little boy, has Asperger's Disease (Syndrom).  It is a side of Autism.
> He is fully functioning, has real social problems and putting pen to paper.
> They are looking for software that would allow him to speak into the
> computer and the computer would then translate the words onto screen for
> him.  Anything you can give me for them would be great.  The young lad is 10
> years old.
>
> Any information you can provide would be greatly appreciated.

ATOM RSS1 RSS2