Hi AndyM,
Take a look at the plot below. What this setup shows is an Earthworks M23R Reference Mic connected to a QA472 Mic Pre with 0 dB of gain and phantom power enabled. That then goes into the QA403. A mic calibrator is connected to the mic. The calibrator has a 94 dB and 114 dB setting. This will show how at least some of the things on your list can be achieved.
The M23R has super tight calibration (+/- 0.5 dB across the band) and a nominal sensitivity of -28.9 dBV/Pa. We can tell the QA40x software the absolute sensitivity of a mic in the dBr context menu (right click on dBR in AXIS control block). Note below we have the -28.9 dBV/Pa sensitivity, and 1 Pa is 94 dB, and thus -28.9 - 94 = -122.9. That is telling the software that when you measure -28.9 dBV, that should display as 94 dB.
Note the gain accuracy on the QA472 is very tight, typically around +/- 0.05 dB It uses an INA849, which is laser trimmed. TI has lots of data on the just how tight that part is. Its like no other mic-pre out there. And the QA403 should get you better than 0.05 dB too. Class 1 is +/- 0.7 dB.
In the plot above, you can see a green reference plot for the 114 dB setting on the calibrator. The live data (blue) is showing for the 94 dBSPL.
The plot below shows 1/3 band response.
There isn’t way to set an integration time and there isn’t a way to measure gapless periods of audio. But, an ASIO driver is available, and so, if there’s software you prefer you could use that. But then you are back to having to calibrate.
There is some other software that has been in the works for doing things like measuring gunshots on a shooting range and live sound/concerts for safety purposes. This will let you continuously log things like A-weighted Leq readings. For example, in the plot below you can see the same left channel as configured above. But this time all the time-averaged metrics you might want are kept and computed. So, you can see where I have clapped near the M23R and hit a peak of 140 dBSPL (red trace is instantaneous peak without any weighting), and then I connected the mic calibrator. Once the calibrator is connected, you’ll see the red trace immediately snap to 94.2 dB SPL and the pink trace (A-weighted Leq over 1 minute) slowly rise to that value over one minute.
This software isn’t ready and I don’t know when/if it will be ready. But it’s a proof of concept of what is possible. And of course, if you can write python or C#, you get the same access to the tightly calibrated audio stream and can do something custom if that makes sense.
