How to effectively use voice recognition
for patient documentation
Published: Dec 10th, 2004
By: John Deutsch
If you don’t already own a copy
of voice recognition software, you probably know someone
who does. Some have had great experiences with it, MOST
have not. Why?
Voice Recognition Software has come a
long way since the early IBM Via Voice and Dragon Naturally
Speaking days. But is it there yet? Is it ready to be
the documentation solution that you’ve been waiting
for all these years? It just might be.
Voice recognition software has been around
for longer than you think. Try 1970. It’s sort
of one of those technologies that everyone has been
waiting for 30 years but it just never really got finished.
Like alternative energy cars. However the advancements
made in this technology over the past 10 years have
been remarkable and more and more products coming to
market these days are equipped with voice activation
ie. Cell Phones, Cars.
So what does this all mean for healthcare
professionals and there endless amounts of paper work?
The technology is mature enough to revolutionize your
documentation, however it is not for the faint of heart.
Do not expect to be able to go to your local computer
store pick up a copy of Dragon Naturally Speaking and
be able to dictate that 10 page report that night. Your
success will be based on these three factors:
Hardware – A powerful computer and
a digital quality microphone (usually USB) is your first
step to starting out on the right foot. Your computer
should be at least a 1 GHz and should have 512MB RAM.
A headset microphone or digital array microphone is
your best bet. If you’re thinking about using
a Tablet PC, Motion Computing’s M1400 has a built-in
microphone that works very well.
Training – The reality 9 out of
10 people will not go home and read the entire users
manual to learn all the features that the software has
to offer. Training from someone who is experienced in
using the software is a great way to go. An hour or
two will go a long way in getting a good voice profile
setup.
Practice – Expect to spend at least
5-10 hours of learning the features, training your profile,
making corrections and getting comfortable with using
it from day to day.
Taking these simple steps should get you
to a level where you will favor voice recognition over
typing, and it’s only up from there. The end result
of the development of this technology will be a universal
profile that will achieve extremely high accuracy with
no training at all, meaning anyone can just pick up
the microphone and use it. This will then allow for
the development of real-time language translators (already
in the works).
John Deutsch is the founder of EMR
Experts www.emrexperts.com, a medical software consulting
firm specializing in EMR software solutions, located
in San Diego, CA. He can be reached at john@emrexperts.com
|
