Hi,You'll need a DC-blocking cap as the microphone pads are designed for an electret capsule. I'm not sure what you mean by a GBF, gain before feedback is a measurement.
Thank you for this precision but my frequency generator does not generate an offset. Now the microphones need bias for their operation, and my generator doesn't need this bias voltage. I'm not sure of my connection.That's a coupling capacitor. The 2.22uF below it is the decoupling cap. The connector is DC-linked to MICBIAS output, which is why an external coupling capacitor needs to be used if the source doesn't handle a DC-offset.
Thank you for your advice, impossible to make a dividing bridge with resistors (too much parasite), I have a -40 dB attenuator but I think the level is a little loud for a microphone input (generally around -60 dB). Probably I'll enter the signal by LINEINL, in this case I can enter a 0 dB level. And of course, check if my software is working well.Most function generators output with a fairly low impedance which can easily drive the input's 2.2K resistor. But to get a signal with low enough amplitude, you might need to add a resistor divider to decrease the signal level.
Indeed the 0.1uF capacitor should block whatever DC level (if any) ends up on the signal. This really should work, though you might end up receiving a clipped signal that's more or less a square wave if the level it too high.