[slightly off-topic] programmable speech recognition software

[joshzeidner at my-deja.com]

> I'm looking for the appropriate SR software to build a part of a room
> installation. The task differs somewhat from the usual "dictate into
> word" applications:

  Is this for a home automation system?

  As far as I can tell... the interface to the SR engine is the least of
your worries.  SR is not widely used... not because it is hard to
program, but because it is unreliable.  The "continuous" SR engines work
partially based on phonemes and partially on sematic contexuality.
Therefore, taking in a stream of utterances and trying to pick a few
words off is impossible.  The chances of recognizing the right word only
comes close to useable if the words to the left or right of that word
are within the right context.  So if you think that you can just feed in
a audio stream and let the engine pick it up- that is not really what SR
can do at this point... Regardless of what IBM or Dragon tells you.  But
it is getting better!!!!

  -Josh Zeidner


>
> In the recorded audio stream should be recognized as many words as
> possible - speaker independent and in German! (I can hear you shudder)
>
> In unclear cases - there will probably be lots of them - the closest
> interpretations would be OK.
>
> Therefore completeness and correctness don't have the highest priority
> in this case.
>
> The application is supposed to run on a Win 9x machine. Direct
> accessibility from Python would be super, otherwise possible via
> COM/SAPI from M$, admittedly I don't know anything about its
> capableness.
>
> I read the postings about NatLink in c.l.p, but in absence of a demo
> version of NatSpeak I cannot tell anything about its suitability, so I
> would be pleased to hear some opinions of this group's numerous SR
> experts!
>
> Regards,
> Stefan
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.