Is there any guidance on how to calibrate the server VAD activation threshold in different environments?
The docs say:
Activation threshold (0 to 1). A higher threshold will require louder audio to activate the model, and thus might perform better in noisy environments.
But there doesn’t appear to be a way to know what the current ambient noise is, in order to set this value appropriately.
For example, I’d like to know if the ambient noise in the room is at a 0.8, so that I can set the activation threshold higher than than that, say a 1.0.
But if the ambient noise in the room is only 0.1, then I should probably set the activation threshold at a 0.3 or something.
Does the activation threshold correspond to a known dB value, or is it more complicated than that?