Sorry, I'm not entirely sure what you are asking here. Based off a couple key words there I'll give it a shot though.

Check out CMU's Sphinx speech recognition engine. This should work on any Windows, Mac or POSIX system (linux, bsd, unix, etc).

For speech output check out CMU's Festival. Again, should work on any OS.

An alternative speech synthesizer is eSpeak. It supposedly runs on Windows or Linux, though I've only used it on the latter. I've had reasonable luck with espeak, it's easy to tweak the voice output and has a decent handling of words (sometimes you have to spell phonetically though).

Watch out, don't step in the anthropomorphization.

To keep it simple, practical, easy to design, I would use a speech recog *chip* (rather than a small *system*, with embedded software; as assembly language in an EPROM or whatever).

I experimented with these, the last time in the mid '90s. They didn't really work that well - you had to *learn* and teach yourself how to speak to them. They are a bit innaccurate. But technology has come a long way.

Google 'speech recognition chip'. Here are a couple of promising links that I found:

http://www.SensoryInc.com

(a kind of review of the above):
http://blog.tmcnet.com/blog/tom-keating/technology-and-science/sensory-speech...

http://www.imagesco.com/articles/speech/speech-recognition-tutorial01.html

is there any voice commands for rs media???????

MrScott, i do belive that is the shortest reply from you I have ever seen.

I started with 2 robots but before I knew it, well, it's a disease I tell you.
RSV1, RSV2, RS Media, Homersapien, Spidersapien, Roboraptor, Robotyrannus and Roboquad

thats messed up

...and AIBO :(

Horia Pernea
http://www.roboticage.eu

FREEEK said: ...and AIBO :(

Which I don't consider to be comparable in price, form, or function.

Too high end.

ScottE -- Member (always) & Moderator (when needed)