Eric wrote:

why but I get significantly worse results when dictating directly into
open office versus something like DragonPad.

For speech recognition both of them require significantly (at least
triple) the stated requrements, for adequte performance.

My suspicion is that open office takes way more resources than simple
editors and this causes problems with how speech recognition programs
read data from the audio stream.

That doesn't only cause problems with speech recognition.  It also
causes problem with speech output.   [This in one reasion why I say
that the minimum acceptable system for running OOo 2.0.x is a 3GHZ
machine with 4GB RAM.  That way you have a fighting chance of getting
acceptable performance.]

ViaVoice the ever helpful monopoly on desktop speech recognition) would
work with open office and make a high accuracy, friendly to use
interface but I'm not holding my breath.

a) What happened to the promise by IBM to opensource ViaVoice?

b) What about using Sphinx?

c) There is a group in Sun that is working on a Java Speech Recognition Engine.

Either b or c should, in theory, be able to either be dropped into
OOo, or work in tandem with OOo.

I think Microsoft would be a better group to work with because they seem to be 
actively reaching out to the handicapped community for not just speech 
recognition but the

You have got to be joking.

The major reason that handicapped "like" Microsoft, is because of the
third party support.
The secondary reason is that the organizations that purchase the
hardware and software neither fund, nor use them.

xan

jonathon
--
Ethical conduct is a vice.
Corrupt conduct is a virtue.

Motto of Nacarima

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to