I am trying to run the speech recognition in the background on Linux machine, but it gives me an AssertionError saying "This audio source is already inside a context ma