I’m working on python automated test framework, and I’m using QA403 for capturing DUT output signal. All is running on ubuntu machine.
I’m using QA40x REST API /Acquisition and /Data/Time/Input which allows me to capture up to 22s~ of audio signal at once with 48k sample rate and max buffer size. My goal is to have long tests and capture 10min of audio or more.
Is there some way to do that, apart from using /Acquisition and /Data/Time/Input in a loop which if I understand correctly will result with loosing some data? I just need to capture raw data in time domain.
Hi @hubert, you should be able to do this with the “bare metal” interface. This gives you complete control of the audio streams with full calibration intact. From this, you can pull hours or days of audio if needed. Demo code in Py, but most AI is really good these days at converting to whatever language you might need.
The lib also shows how to handle things like normalization so that your FFT results make sense. That is, if you sample a 1Vrms time series, apply a window and convert to freq domain via FFT, the FFT will show a peak of 1Vrms.
Thanks @matt! I managed to use the lib with small modifications. I would like to potentially package forked version for my own workflow, but I noticed there is no licence added in the original repo. Would you be open to adding an open-source license to the repository?