here are updated patches based on the review. Also I manged to fix the issues in the sanity-check phase.
However, there are still some points to be solved
as the ALSA lib error blocks the tests. I also don't know how to mock the audio subsystem in tests properly.
FIY I renamed the package from python-mycroft-core to mycroft-core and move it to gnu/packages/speech.scm as this IMHO reflect more the purpose of this package.
Python module xdg was added as it's referenced in the code.