The Right Stuff (Part 2)

In my previous entry, I began to discuss what I believe are features
critical to an augmented communication system (ACS) that expands the
capabilities of its users rather than becoming just another obstacle for them
to overcome. I make no claim of having a degree in SLP or any related
discipline. All I have in sheepskin is a bachelor of science in Computing
Science. No, my perspective is that of a so-called severely disabled person who
gained and held an application developer for a major oil company for 14 years. Although I was supported and encouraged by my employer
in trying several existing ACS, my managers and co-workers found none of them
good enough to carry on a conversation with me.
That’s what
drove me to create my own solution. While I didn’t get as many opportunities
to participate at Shell as I would have liked, there was no doubt that the
number significantly improved after I developed Xpress It.


In essence, there are three qualities necessary for an ACS to be a net
asset for its user outside of the carefully controlled environments where such
things are typically evaluated.:


  • Intelligibility:
    this is the most important element of all. An ACS may do
    a wonderful job of predicting entire phrases from quite minimal user input,
    but it will fail to help the person out in “the real world” if
    complete strangers find it difficult to understand. I make no secret of
    the fact that my solution employs the Eloquent Technologies’ speech
    engine. Before ETI came along, the best answer was a computer peripheral
    called DECTalk. DECTalk was actually fairly credible with diction, but its
    innate ability to correctly pronounce exceptional words and provide at
    least adequate inflection just from sentence structure was underdeveloped.
    ETI’s engine, while a huge leap forward, is still a work in progress.
     We know that people commonly lose portions of the audible spectrum
    as they age. Older people also tend to be management. They partly
    compensate for any hearing deficit by inferring meaning from inflection. That’s
    why simply having a good voice synthesizer isn’t enough to really
    help the user. It must be drop-dead gorgeous if that user hopes to be accepted
    in the community.
  • Flexibility:
    one criteria speech pathologists use in evaluating an
    ACS is its ability to predict as much as possible. I’ve previously acknowledged
    that there is a reasonable logic to this. Yet, it seems best suited to
    users with limited mental function. Any disabled person trying to survive
    in the business world must be every bit as eloquent as his/her ACS can
    deal with. As such, I’ve found prediction beyond the next word to be
    a dubious feature at best. An accidental selection can lead to more time
    lost than a successful prediction could hope to gain. My ACS does have the
    ability to assign mnemonics to a virtually unlimited number of phrases. Yet,
    I’ve only found one really useful. That’s the three sentences I
    use to greet strangers over the phone and explain the simple rules that allow
    us to conduct a telephone conversation. After that, the discussion is free
    to go wherever it needs to.
  • Compatibility:
    we do not typically converse in a vacuum. That is, we
    converse usually while performing other tasks. That makes it a crime for
    an ACS to interfere in any way in performing other tasks. There’s
    simply no excuse for that, but all existing ACS appear to be quite guilty
    of it in the first degree. If we are discussing computer-based solutions,
    there’s no technical reason for an ACS to “take over” or
    otherwise inhibit a computer from being concurrently used for other work. Just
    view the situation from a businessman’s perspective. Why should he
    hire someone who must stop everything they’re doing to ask or answer
    a simple question? He shouldn’t, and likely won’t. Then why is
    my ACS the only one that’s transparent to its host computer?


As I’ve said before, it really isn’t the goal of this blog
to market my solution. True, generating some sales currently seems my only path
to continued economic survival. However, I created this blog to be another tool
useful to challenge—and frankly embarrass—the ACS industry into
starting innovations that were possible 15 years ago! In part, that unfortunately
means speech pathologists must somehow find the time to become more computer-aware.
I understand how difficult that is, but how else do you learn to recognize when
a vendor is trying to sell you technology that should’ve died with DOS? Innovation
will only happen when those empowered to make recommendations start raising
their expectations. If I were the parent of a speech-impaired child, I’d
have a lot of sleepless nights worrying about his/her future. On one hand, there
is little right now to give cause for much hope for the future. Disabled people
are discriminated against to a magnitude that I can only hope future generations
won’t have to contend with. On the other hand, the future is full of bright
possibilities, including perhaps a PDA-based version of Xpress It to put GUS to
shame. However, getting there requires using the very best that is available
today in order to power development of that brighter future.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s