Our experience with voice recognition is that it still needs a very closely
coupled microphone of known quality.  Ambient background 
and noise messes it up thoroughly.  

Although this kind of device could be of help, it isn't really a "solution"
because even though you know what you told it to do, you don't know if it 
understood properly.  By that I mean the entire system.  We have to keep it 
understood that any really good accessibility system demands the same level
of feedback that sightlings get.  You gotta know what the machine you're running
is up to, not just what you told it to be up too.

Just a rehab engineer's diatribe

I too have and do put up with devices that are just barely runnable sometimes
because of a lack of properfeedback, but I don't love them.

Like that "video level" from Zercon, If I really needed it, I'd hate it.

diatribe ends, for now! <GRIN>
Tom

Reply via email to