On 9/27/06, Paul G. Allen <[EMAIL PROTECTED]> wrote:
Rick Funderburg wrote:
> [EMAIL PROTECTED] wrote:
>
>
>> Perhaps speech recognition may be the answer
>> some day but I don't see it now.
>>
>> Is there any other alternative to voice to replace
>> the keyboard?
>>
>>
>
>
> Speech recognition should get somewhat better, but there are limitations
> when grabbing samples of air vibrations. Humans compensate for this by
> analyzing the context, which is something that is very hard for
> computers to do. I assume that one could get much better accuracy by
> implanting small sensors into the muscles that are involved in speech.
> I have had a similar thought for a version of sign language using some
> sort of glove. These ideas may help until a direct-to-brain interface
> is working.
>
Ever hear of VoiceType(R)? It is speech recognition software that came with
OS/2 Warp 4.
It was excellent in that you could navigate the computer and write entire
documents
without ever touching the keyboard. It was also trainable - if it didn't know a
word, then
you could teach it. The drawback was it took a good deal of memory and some
processing
power, but if used today on faster machines, You'd hardly notice the resources
it used.
Memory is getting smaller (physically) and faster as well, so large memory and
storage
should not be an issue in embedded systems as it once was. Note that by
comparison to
today's application standards, VoiceType(R) took hardly any memory or storage
at all.
Just to put some kind of historical perspective on this, when I was at
the International Congress on Acoustics in 1956 (half a century ago,
wow!) automatic voice recognition hardware was "just around the
corner". It is still a long corner.
carl
--
carl lowenstein marine physical lab u.c. san diego
[EMAIL PROTECTED]
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
