(Sorry if you receive this mail twice. I've sent it the past Friday but I
beleave it got lost... so I'm sending it again, not from a mail.ru account.)

Stephen Travis Pope wrote:
>
> Does anybody out there have any plans for porting some kind of speech
> *input* framework? I know that there was a Smalltalk interface to the
> Chant speech recognizer, and an interface to Sphinx a while back...
>

I start working on that some time ago, but I didn't find the time to
finish it. What I have done is an LPAnalyzer (and the corresponding
LPSynthesizer)... the next step would be to implement a DTW algorithm,
which is pretty simple I beleave. That could be enough for recognition
if isolated words.

Luciano.-