what would you programmers out there think about starting a self-made voice-recognition project?

would there be a possibility to do something like that?
someone out there who would be able to do that?
any ideas?

I would be interested, but I cannot code my way out of a paper bag. One approach, though: it does not have to be a complicated interface. All you need to recognize are five keyworks: up, down, left, right and knob. Well, maybe six if you want to have long presses... Perhaps a good way to start is to have six wave (or mp3, but I don't know about the compression for this application) files stored that the empeg could compare against...

I also wouldn't have enough knowledge of the Empeg or the Linux platform to try this yet, but I'd love to see it happen. I had a quick look around and found this open source VR project, which may also provide a starting point.
http://sourceforge.net/projects/cmusphinx/

I do like the idea of initially using VR to mimick the simple button-presses.

If we wanted to get really basic then we could get the user to whistle tones to control the empeg. It shouldn't be too hard to make it work as well. Much simpler than VR anyway...

- Trevor