RFE: use microphone to sample ambient loudness

[Lee-Carre]

Based on #27, since the goal is to compensate for the louder noise which results from speed, then (optionally) measure the noise directly rather than only assuming indirectly via speed.

This would make Speed of Sound more adaptive & responsive, to more situations, including edge-cases.

Some vehicles have a very non-linear speed-versus-loudness profile. Aircraft & trains come to mind. Though, I recognise that the original / intended use-case is road vehicles.