Mic set up for Speech Recognition - Help!

[Smelter]

I have a fully working audio set-up but Windows Speech Recognition fails to hear the microphone.
To clarify the situation. Under Windows 7 64bit I have an E-MU 1616 audio unit that uses the "PatchMix DSP" software which is a virtual sound mixing desk.

The Microphone works.
PatchMix 'hears' and shows on the bargraph meters.
I hear it on the monitor speakers.
SoundForge (audio editor) hears it and also shows on the bargraph meters and 'records' it.
So I know it all works through the ASIO drivers.

I learnt through various forum postings about the missing 'ctfmon.exe' which now runs and I get past the "optimized for speakers of English (UK)" screen.
Sadly selecting any of the three microphone types offered, none show up on the bargraph of the 'Adjust volume' page.

Under Windows 'Sound' :-
Playback Speakers show correctly on the bargraph for the default.
Recording setting has "E-MU E-DSP Audio Processor (WDM)" as the default device, but no bargraph presence.
I checked Windows sound scheme does work, but it is set to 'No Sounds'

So anyone have any suggestions for what else I can do to get my mic set up for Speech Recognition.

[iceblast]

You might want to try with all those programs turned off. It could be interfering with Windows Speech Recognition from hearing the mic.

[Smelter]

You might want to try with all those programs turned off. It could be interfering with Windows Speech Recognition from hearing the mic.

Sadly that was one of the first avenues I tried and failed.
I am beginning to think that the "Microphone Setup Wizard" doesn't understand ASIO channels, and hence the "Microsoft Speech Recognition" won't let me get to the set-up menu.

[Smelter]

Well I now have a working solution, but not what I had set out to use.
It turns out that pro audio outboard mixers don't talk to Windows through the standard audio path of mic and speakers. Yes they will receive "wave" file format data on an input channel but talk back to applications via ASIO channels.
So the solution was to use the freebie USB audio interface and cable thrown in with my new condenser microphone which unbalances the mic's output and does not supply phantom power. The lack of power to the mic results in a very low level signal but speaking close up to the mic means that the speach recognition software works and it does not pick up the studio monitor speakers.
Of course the mic with phantom power and a balanced output via a true XLR cable into the mixer is far superior, but hey I now have a working setups for both uses.
Winner.