Hi id,
Getting a USB audio device to work on a cRIO is going to be non-trivial. The reasons for that have already been discussed over here: http://forums.ni.com/t5/NI-Linux-Real-Time-Discussions/Using-USB-audio-device-on-cRIO-9030-and-cRIO-...
That said, I'm curious- why the cRIO platform? Are you doing other things alongside the speech recognition that utilize the C Series chassis? Or are you hoping to utilize the FPGA in doing your speech recognition algorithm? If you are not using anything else from the C Series lineup and simply want the FPGA processing, you might look at the myRIO instead, since that has 3.5 mm jacks on it, is much less expensive, and has approximately equivalent processing power as the cRIO-906x family.
Now, it's worth noting that you have plenty of options for hooking up a microphone or speaker to a cRIO. It's just not going to be a plug-and-play 3.5 mm jack, and as far as the cRIO knows, it's just an analog voltage signal. You will probably need to use a preamplifier, since an average dynamic microphone is going to output in the 1-100 uV range, while most of our C Series DAQ devices are going to read around 10 uV/bit at best. From there, you just sample analog voltage, and decide what exactly each sample means.
William R.
