In my old job we used a lot of Shure wireless, and I have personal experience with the PGX series wireless mics. They work well at shorter ranges (which 25x25 fits in easily), and are easy to set up. A system with lavalier mic is about $500. I preferred the WL185 to the WL93. It's bigger and a little more sensitive.
The one problem you may have is the PGX does not have removable antennas, so you may need to mount the entire receiver in the open. If you need something with removable antennas, you're looking at hundreds more if you stick with Shure.
There are otheres, like Audio-Technica or even Samson which may work, but I can't comment on their reliability. I've used some older cheaper wireless mics and interference is a realistic issue, as is battery life. They just aren't built the same.
You shouldn't have any issue connecting the output of the mic to the P8221 but you will most likely need to adapt an XLR cable to a stereo plug or screw terminals. Not hard, just outside the realm of surveillance most are used to!
There are a number of variables of course. Machinery locations, where throughout the room that conversations are occurring (ie... around a table, workbench, or at multiple work stations). How many people at one time. Wall, flooring and ceiling materials, etc...
While not wireless, it is possible a Boundary Microphone may do the job. A Crown PZM mic is among the best at picking up sound in a variety of room settings. You would still use the Axis P8221 connected to an XLR to 3.5mm adapter for many of these type mics as mentioned by Ethan.
If your mic output is a balanced one (most likely a 3-pin XLR male) make sure to use audio input 2, (the screw terminal one) since it's the balanced one. Also apply the output gain (if any) of the wireless receiver first before increasing that of the axis module to maximize s/n ratio.
To increase speech clarity and reduce background noise you may want to try a lav with a cardioid (unidirectional) pickup pattern as opposed to an omnidirectional one. Beware though if the speaker's head turns much this can cause a drop in volume.
The ideal solution from an audio quality standpoint is a headset mic, though this is also the most restrictive to the wearer.
Idibri Consulting | IPVMU Certified | 09/18/14 10:25pm
I did something similar to what you describe a few years ago. It was for a restaurant chain that had several test/training kitchens in their corporate offices. I used a Lectrosonic wireless mic kit intended for ENG (electronic news gathering). It uses a beltpack transmitter and a beltpack receiver that can be mounted to the camera.
The receiver model I used is discontinued now, but it was similar to this model. I used an Axis Q1755 camera, which was the only HD security camera at the time. I just secured the receivers to the cameras with Velcro and connected the audio out of the receiver to the external mic input on the camera. The receiver had an output level adjust so that I could turn it down. Otherwise, it would overload the mic input of the Q1755. It was also able to be powered by external DC power, rather than batteries. That was important because the cameras were ceiling mounted and they did not want to have to climb up to them to turn the mic receivers on/off or change batteries.
It worked reasonably well. The biggest problem I had at that time was that the only VMS that I could find that synchronized audio with video was DVTEL Latitude. With all of the others I tested, (every VMS I could find that would give me a demo program to test), the audio and video were horribly out of sync. Some of them by 2 seconds or more. Hopefully they've gotten better by now.