Personal Audio Classifier extension error

I am trying to create a voice authentication app by following the tutorial at: http://appinventor.mit.edu/assets/files/voice-authentication-app-tutorial.pdf
I confirmed that my phone is compatible with the PAC extension by running the TestPAC app.

I was able to successfully train the voice model and create an APK with the PAC extension. However, when I try authenticating using my voice through the app in my Android phone, PAC returns the following error: "Failed to create spectrogram: TypeError: Failed to fetch". I have tried this across multiple Android devices (OnePlus 3T, Google Pixel 4a, Samsung S6 lite tablet) and all return the same error (all of these devices are compatible with PAC as per the TestPAC app).

I looked at How to trigger classification in Personal Audio Classifier extension? and noticed that the PAC extension is already the latest version (Extension Version: 20200904).

Screenshot:

Any help would be appreciated.

I too use that extension and it works fine for me. Can you give the aia file so that we can see if it is problem with your code or your device ?